feat(route): add Cool Papers #14129

nczitzk · 2023-12-28T15:58:36Z

Involved Issue / 该 PR 相关 Issue

Example for the Proposed Route(s) / 路由地址示例

/papers/arxiv/cs.CL

New RSS Route Checklist / 新 RSS 路由检查表

Note / 说明

github-actions · 2023-12-28T16:02:32Z

Successfully generated as following:

http://localhost:1200/papers/arxiv/cs.CL - Success ✔️

<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"
>
    <channel>
        <title><![CDATA[Computation and Language | Cool Papers]]></title>
        <link>https://papers.cool/arxiv/cs.CL</link>
        <atom:link href="http://localhost:1200/papers/arxiv/cs.CL" rel="self" type="application/rss+xml" />
        <description><![CDATA[Computation and Language | Cool Papers - Immersive Paper Discovery - Made with love by RSSHub(https://github.com/DIYgod/RSSHub)]]></description>
        <generator>RSSHub</generator>
        <webMaster>i@diygod.me (DIYgod)</webMaster>
        <language>zh-cn</language>
        <lastBuildDate>Thu, 28 Dec 2023 16:02:31 GMT</lastBuildDate>
        <ttl>5</ttl>
        <item>
            <title><![CDATA[Principled Instructions Are All You Need for Questioning LLaMA-1/2, GPT-3.5/4]]></title>
            <description><![CDATA[<p>This paper introduces 26 guiding principles designed to streamline the process of querying and prompting large language models. Our goal is to simplify the underlying concepts of formulating questions for various scales of large language models, examining their abilities, and enhancing user comprehension on the behaviors of different scales of large language models when feeding into different prompts. Extensive experiments are conducted on LLaMA-1/2 (7B, 13B and 70B), GPT-3.5/4 to verify the effectiveness of the proposed principles on instructions and prompts design. We hope that this work provides a better guide for researchers working on the prompting of large language models. Project page is available at https://github.com/VILA-Lab/ATLAS.</p>
  <p class="faq-q"><strong>Q</strong>: 这篇论文试图解决什么问题？</p><p class="faq-a"><strong>A</strong>:  这篇论文旨在简化查询和提示大型语言模型（LLMs）的过程，提出了26个指导原则，以增强用户对不同规模LLMs行为的理解。这些原则旨在改善LLMs的提示设计，提高LLMs在各种任务中的表现，尤其是在生成问题答案时。论文的目标是为研究人员提供更好的指导，以便在提示LLMs时能够获得更高质量的响应。</p><p class="faq-q"><strong>Q</strong>: 有哪些相关研究？</p><p class="faq-a"><strong>A</strong>:  这篇论文提到了以下相关研究：
1. 语言模型的演变，如BERT、T5、GPT-1、GPT-2、GPT-3、Gopher、LLaMA系列和Mistral。
2. 提示工程，包括Ask-Me-Anything提示、Chain-of-Thought方法、least-to-most提示、解释性提示、few-shot提示、输出引导提示等。
3. 提示工程技术目录，强调了在软件开发和教育中提高LLM性能的重要性。
4. 方向性刺激提示，提出了一种使用可调策略模型生成辅助提示的新框架，指导LLMs实现特定期望结果。</p><p class="faq-q"><strong>Q</strong>: 论文如何解决这个问题？</p><p class="faq-a"><strong>A</strong>:  论文通过以下方式解决这个问题：
1. 提出了26个原则，以改善LLMs的提示质量。这些原则涵盖了提示结构和清晰度、特定性、用户互动和参与、内容和语言风格以及复杂任务和编码提示等方面。
2. 在ATLAS基准测试上进行了实验，这是一个手动制作的基准测试，包含针对每个原则的多个问题。实验结果表明，与标准提示相比，应用这些原则的提示可以显著提高LLMs响应的质量和准确性。
3. 分析了不同规模的LLMs（如LLaMA-1/2和GPT-3.5/4）在应用这些原则时的表现。结果表明，随着模型规模的增加，性能提升更加明显。</p><p class="faq-q"><strong>Q</strong>: 论文做了哪些实验？</p><p class="faq-a"><strong>A</strong>:  论文在ATLAS基准测试上进行了实验，这是一个手动制作的基准测试，包含针对每个原则的多个问题。实验分为两个部分：提升和正确性。通过人类评估来衡量LLMs输出的质量提升和准确性。实验使用了以下模型：LLaMA-1-{7, 13}, LLaMA-2-{7, 13}, LLaMA-2-70B-chat, GPT-3.5 (ChatGPT) 和 GPT-4。这些模型分为不同的规模：小型（7B模型）、中型（13B）和大型（70B, GPT-3.5/4）。实验结果表明，应用这些原则的提示可以显著提高LLMs响应的质量和准确性。</p><p class="faq-q"><strong>Q</strong>: 有什么可以进一步探索的点？</p><p class="faq-a"><strong>A</strong>:  论文提出了以下未来探索方向：
1. 使用替代策略（如微调、强化学习、直接偏好优化或使用生成的数据集的不同提示方法）来进一步优化基础模型，使其更好地符合原则性指导。
2. 将成功的策略整合到标准的LLM操作中，例如使用原始/原则性提示作为输入，经过优化的原则性响应作为训练目标进行微调。
3. 在未来的研究中扩展问题集，以获得更普遍的发现并深入了解每个原则的适用性。</p><p class="faq-q"><strong>Q</strong>: 总结一下论文的主要内容</p><p class="faq-a"><strong>A</strong>:  这篇论文提出了26个原则，旨在改善大型语言模型（LLMs）在回答问题时的表现。这些原则涵盖了提示的结构和清晰度、特定性、用户互动和参与、内容和语言风格以及复杂任务和编码提示等方面。论文在ATLAS基准测试上进行了实验，评估了不同规模的LLMs（如LLaMA-1/2和GPT-3.5/4）在应用这些原则时的表现。实验结果表明，应用这些原则的提示可以显著提高LLMs响应的质量和准确性。此外，论文还讨论了未来的研究方向，如使用替代策略进一步优化模型，将成功的策略整合到标准的LLM操作中，以及扩展问题集以获得更普遍的发现。</p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16171?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16171</link>
            <enclosure url="https://arxiv.org/pdf/2312.16171"  type="application/pdf" />
            <author><![CDATA[Sondos Mahmoud Bsharat; Aidar Myrzakhan; Zhiqiang Shen]]></author>
        </item>
        <item>
            <title><![CDATA[From Text to Multimodal: A Comprehensive Survey of Adversarial Example Generation in Question Answering Systems]]></title>
            <description><![CDATA[<p>Integrating adversarial machine learning with Question Answering (QA) systems has emerged as a critical area for understanding the vulnerabilities and robustness of these systems. This article aims to comprehensively review adversarial example-generation techniques in the QA field, including textual and multimodal contexts. We examine the techniques employed through systematic categorization, providing a comprehensive, structured review. Beginning with an overview of traditional QA models, we traverse the adversarial example generation by exploring rule-based perturbations and advanced generative models. We then extend our research to include multimodal QA systems, analyze them across various methods, and examine generative models, seq2seq architectures, and hybrid methodologies. Our research grows to different defense strategies, adversarial datasets, and evaluation metrics and illustrates the comprehensive literature on adversarial QA. Finally, the paper considers the future landscape of adversarial question generation, highlighting potential research directions that can advance textual and multimodal QA systems in the context of adversarial challenges.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16156?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16156</link>
            <enclosure url="https://arxiv.org/pdf/2312.16156"  type="application/pdf" />
            <author><![CDATA[Gulsum Yigit; Mehmet Fatih Amasyali]]></author>
        </item>
        <item>
            <title><![CDATA[The Media Bias Taxonomy: A Systematic Literature Review on the Forms and Automated Detection of Media Bias]]></title>
            <description><![CDATA[<p>The way the media presents events can significantly affect public perception, which in turn can alter people's beliefs and views. Media bias describes a one-sided or polarizing perspective on a topic. This article summarizes the research on computational methods to detect media bias by systematically reviewing 3140 research papers published between 2019 and 2022. To structure our review and support a mutual understanding of bias across research domains, we introduce the Media Bias Taxonomy, which provides a coherent overview of the current state of research on media bias from different perspectives. We show that media bias detection is a highly active research field, in which transformer-based classification approaches have led to significant improvements in recent years. These improvements include higher classification accuracy and the ability to detect more fine-granular types of bias. However, we have identified a lack of interdisciplinarity in existing projects, and a need for more awareness of the various types of media bias to support methodologically thorough performance evaluations of media bias detection systems. Concluding from our analysis, we see the integration of recent machine learning advancements with reliable and diverse bias assessment strategies from other research areas as the most promising area for future research contributions in the field.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16148?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16148</link>
            <enclosure url="https://arxiv.org/pdf/2312.16148"  type="application/pdf" />
            <author><![CDATA[Timo Spinde; Smilla Hinterreiter; Fabian Haak; Terry Ruas; Helge Giese; Norman Meuschke; Bela Gipp]]></author>
        </item>
        <item>
            <title><![CDATA[JaColBERT and Hard Negatives, Towards Better Japanese-First Embeddings for Retrieval: Early Technical Report]]></title>
            <description><![CDATA[<p>Document retrieval in many languages has been largely relying on multi-lingual models, and leveraging the vast wealth of English training data. In Japanese, the best performing deep-learning based retrieval approaches rely on multilingual dense embeddings. In this work, we introduce (1) a hard-negative augmented version of the Japanese MMARCO dataset and (2) JaColBERT, a document retrieval model built on the ColBERT model architecture, specifically for Japanese. JaColBERT vastly outperform all previous monolingual retrieval approaches and competes with the best multilingual methods, despite unfavourable evaluation settings (out-of-domain vs. in-domain for the multilingual models). JaColBERT reaches an average Recall@10 of 0.813, noticeably ahead of the previous monolingual best-performing model (0.716) and only slightly behind multilingual-e5-base (0.820), though more noticeably behind multilingual-e5-large (0.856). These results are achieved using only a limited, entirely Japanese, training set, more than two orders of magnitudes smaller than multilingual embedding models. We believe these results show great promise to support retrieval-enhanced application pipelines in a wide variety of domains.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16144?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16144</link>
            <enclosure url="https://arxiv.org/pdf/2312.16144"  type="application/pdf" />
            <author><![CDATA[Benjamin Clavié]]></author>
        </item>
        <item>
            <title><![CDATA[RoleEval: A Bilingual Role Evaluation Benchmark for Large Language Models]]></title>
            <description><![CDATA[<p>The rapid evolution of large language models (LLMs) necessitates effective benchmarks for evaluating their role knowledge, which is essential for establishing connections with the real world and providing more immersive interactions. This paper introduces RoleEval, a bilingual benchmark designed to assess the memorization, utilization, and reasoning capabilities of role knowledge. RoleEval comprises RoleEval-Global (including internationally recognized characters) and RoleEval-Chinese (including characters popular in China), with 6,000 Chinese-English parallel multiple-choice questions focusing on 300 influential people and fictional characters drawn from a variety of domains including celebrities, anime, comics, movies, TV series, games, and fiction. These questions cover basic knowledge and multi-hop reasoning abilities, aiming to systematically probe various aspects such as personal information, relationships, abilities, and experiences of the characters. To maintain high standards, we perform a hybrid quality check process combining automatic and human verification, ensuring that the questions are diverse, challenging, and discriminative. Our extensive evaluations of RoleEval across various open-source and proprietary large language models, under both the zero- and few-shot settings, reveal insightful findings. Notably, while GPT-4 outperforms other models on RoleEval-Global, Chinese LLMs excel on RoleEval-Chinese, highlighting significant knowledge distribution differences. We expect that RoleEval will highlight the significance of assessing role knowledge for foundation models across various languages and cultural settings.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16132?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16132</link>
            <enclosure url="https://arxiv.org/pdf/2312.16132"  type="application/pdf" />
            <author><![CDATA[Tianhao Shen; Sun Li; Deyi Xiong]]></author>
        </item>
        <item>
            <title><![CDATA[Dotless Representation of Arabic Text: Analysis and Modeling]]></title>
            <description><![CDATA[<p>This paper presents a novel dotless representation of Arabic text as an alternative to the standard Arabic text representation. We delve into its implications through comprehensive analysis across five diverse corpora and four different tokenization techniques. We explore the impact of dotless representation on the relationships between tokenization granularity and vocabulary size and compare them with standard text representation. Moreover, we analyze the information density of dotless versus standard text using text entropy calculations. To delve deeper into the implications of the dotless representation, statistical and neural language models are constructed using the various text corpora and tokenization techniques. A comparative assessment is then made against language models developed using the standard Arabic text representation. This multifaceted analysis provides valuable insights into the potential advantages and challenges associated with the dotless representation. Last but not the least, utilizing parallel corpora, we draw comparisons between the text analysis of Arabic and English to gain further insights. Our findings shed light on the potential benefits of dotless representation for various NLP tasks, paving the way for further exploration for Arabic natural language processing.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16104?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16104</link>
            <enclosure url="https://arxiv.org/pdf/2312.16104"  type="application/pdf" />
            <author><![CDATA[Maged S. Al-Shaibani; Irfan Ahmad]]></author>
        </item>
        <item>
            <title><![CDATA[A Logically Consistent Chain-of-Thought Approach for Stance Detection]]></title>
            <description><![CDATA[<p>Zero-shot stance detection (ZSSD) aims to detect stances toward unseen targets. Incorporating background knowledge to enhance transferability between seen and unseen targets constitutes the primary approach of ZSSD. However, these methods often struggle with a knowledge-task disconnect and lack logical consistency in their predictions. To address these issues, we introduce a novel approach named Logically Consistent Chain-of-Thought (LC-CoT) for ZSSD, which improves stance detection by ensuring relevant and logically sound knowledge extraction. LC-CoT employs a three-step process. Initially, it assesses whether supplementary external knowledge is necessary. Subsequently, it uses API calls to retrieve this knowledge, which can be processed by a separate LLM. Finally, a manual exemplar guides the LLM to infer stance categories, using an if-then logical structure to maintain relevance and logical coherence. This structured approach to eliciting background knowledge enhances the model's capability, outperforming traditional supervised methods without relying on labeled data.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16054?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16054</link>
            <enclosure url="https://arxiv.org/pdf/2312.16054"  type="application/pdf" />
            <author><![CDATA[Bowen Zhang; Daijun Ding; Liwen Jing; Hu Huang]]></author>
        </item>
        <item>
            <title><![CDATA[DocMSU: A Comprehensive Benchmark for Document-level Multimodal Sarcasm Understanding]]></title>
            <description><![CDATA[<p>Multimodal Sarcasm Understanding (MSU) has a wide range of applications in the news field such as public opinion analysis and forgery detection. However, existing MSU benchmarks and approaches usually focus on sentence-level MSU. In document-level news, sarcasm clues are sparse or small and are often concealed in long text. Moreover, compared to sentence-level comments like tweets, which mainly focus on only a few trends or hot topics (e.g., sports events), content in the news is considerably diverse. Models created for sentence-level MSU may fail to capture sarcasm clues in document-level news. To fill this gap, we present a comprehensive benchmark for Document-level Multimodal Sarcasm Understanding (DocMSU). Our dataset contains 102,588 pieces of news with text-image pairs, covering 9 diverse topics such as health, business, etc. The proposed large-scale and diverse DocMSU significantly facilitates the research of document-level MSU in real-world scenarios. To take on the new challenges posed by DocMSU, we introduce a fine-grained sarcasm comprehension method to properly align the pixel-level image features with word-level textual features in documents. Experiments demonstrate the effectiveness of our method, showing that it can serve as a baseline approach to the challenging DocMSU. Our code and dataset are available at https://github.com/Dulpy/DocMSU.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16023?kiwi</guid>
            <link>https://arxiv.org/abs/2312.16023</link>
            <enclosure url="https://arxiv.org/pdf/2312.16023"  type="application/pdf" />
            <author><![CDATA[Hang Du; Guoshun Nan; Sicheng Zhang; Binzhu Xie; Junrui Xu; Hehe Fan; Qimei Cui; Xiaofeng Tao; Xudong Jiang]]></author>
        </item>
        <item>
            <title><![CDATA[Aligning Large Language Models with Human Preferences through Representation Engineering]]></title>
            <description><![CDATA[<p>Aligning large language models (LLMs) with human preferences is crucial for enhancing their utility in terms of helpfulness, truthfulness, safety, harmlessness, and interestingness. Existing methods for achieving this alignment often involves employing reinforcement learning from human feedback (RLHF) to fine-tune LLMs based on human labels assessing the relative quality of model responses. Nevertheless, RLHF is susceptible to instability during fine-tuning and presents challenges in implementation.Drawing inspiration from the emerging field of representation engineering (RepE), this study aims to identify relevant representations for high-level human preferences embedded in patterns of activity within an LLM, and achieve precise control of model behavior by transforming its representations. This novel approach, denoted as Representation Alignment from Human Feedback (RAHF), proves to be effective, computationally efficient, and easy to implement.Extensive experiments demonstrate the efficacy of RAHF in not only capturing but also manipulating representations to align with a broad spectrum of human preferences or values, rather than being confined to a singular concept or function (e.g. honesty or bias). RAHF's versatility in accommodating diverse human preferences shows its potential for advancing LLM performance.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15997?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15997</link>
            <enclosure url="https://arxiv.org/pdf/2312.15997"  type="application/pdf" />
            <author><![CDATA[Wenhao Liu; Xiaohua Wang; Muling Wu; Tianlong Li; Changze Lv; Zixuan Ling; Jianhao Zhu; Cenyuan Zhang; Xiaoqing Zheng; Xuanjing Huang]]></author>
        </item>
        <item>
            <title><![CDATA[Towards Probing Contact Center Large Language Models]]></title>
            <description><![CDATA[<p>Fine-tuning large language models (LLMs) with domain-specific instructions has emerged as an effective method to enhance their domain-specific understanding. Yet, there is limited work that examines the core characteristics acquired during this process. In this study, we benchmark the fundamental characteristics learned by contact-center (CC) specific instruction fine-tuned LLMs with out-of-the-box (OOB) LLMs via probing tasks encompassing conversational, channel, and automatic speech recognition (ASR) properties. We explore different LLM architectures (Flan-T5 and Llama), sizes (3B, 7B, 11B, 13B), and fine-tuning paradigms (full fine-tuning vs PEFT). Our findings reveal remarkable effectiveness of CC-LLMs on the in-domain downstream tasks, with improvement in response acceptability by over 48% compared to OOB-LLMs. Additionally, we compare the performance of OOB-LLMs and CC-LLMs on the widely used SentEval dataset, and assess their capabilities in terms of surface, syntactic, and semantic information through probing tasks. Intriguingly, we note a relatively consistent performance of probing classifiers on the set of probing tasks. Our observations indicate that CC-LLMs, while outperforming their out-of-the-box counterparts, exhibit a tendency to rely less on encoding surface, syntactic, and semantic properties, highlighting the intricate interplay between domain-specific adaptation and probing task performance opening up opportunities to explore behavior of fine-tuned language models in specialized contexts.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15922?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15922</link>
            <enclosure url="https://arxiv.org/pdf/2312.15922"  type="application/pdf" />
            <author><![CDATA[Varun Nathan; Ayush Kumar; Digvijay Ingle; Jithendra Vepa]]></author>
        </item>
        <item>
            <title><![CDATA[Supervised Knowledge Makes Large Language Models Better In-context Learners]]></title>
            <description><![CDATA[<p>Large Language Models (LLMs) exhibit emerging in-context learning abilities through prompt engineering. The recent progress in large-scale generative models has further expanded their use in real-world language applications. However, the critical challenge of improving the generalizability and factuality of LLMs in natural language understanding and question answering remains under-explored. While previous in-context learning research has focused on enhancing models to adhere to users' specific instructions and quality expectations, and to avoid undesired outputs, little to no work has explored the use of task-Specific fine-tuned Language Models (SLMs) to improve LLMs' in-context learning during the inference stage. Our primary contribution is the establishment of a simple yet effective framework that enhances the reliability of LLMs as it: 1) generalizes out-of-distribution data, 2) elucidates how LLMs benefit from discriminative models, and 3) minimizes hallucinations in generative tasks. Using our proposed plug-in method, enhanced versions of Llama 2 and ChatGPT surpass their original versions regarding generalizability and factuality. We offer a comprehensive suite of resources, including 16 curated datasets, prompts, model checkpoints, and LLM outputs across 9 distinct tasks. Our empirical analysis sheds light on the advantages of incorporating discriminative models into LLMs and highlights the potential of our methodology in fostering more reliable LLMs.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15918?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15918</link>
            <enclosure url="https://arxiv.org/pdf/2312.15918"  type="application/pdf" />
            <author><![CDATA[Linyi Yang; Shuibai Zhang; Zhuohao Yu; Guangsheng Bao; Yidong Wang; Jindong Wang; Ruochen Xu; Wei Ye; Xing Xie; Weizhu Chen; Yue Zhang]]></author>
        </item>
        <item>
            <title><![CDATA[Align on the Fly: Adapting Chatbot Behavior to Established Norms]]></title>
            <description><![CDATA[<p>In this paper, we aim to align large language models with the ever-changing, complex, and diverse human values (e.g., social norms) across time and locations. This presents a challenge to existing alignment techniques, such as supervised fine-tuning, which internalize values within model parameters. To overcome this, we propose an On-the-fly Preference Optimization (OPO) method, which is a real-time alignment that works in a streaming way. It employs an external memory to store established rules for alignment, which can constrain LLMs' behaviors without further training, allowing for convenient updates and customization of human values. We also introduce a scalable evaluation to assess the proposed method more effectively. Experimental results on both human-annotated and auto-generated questions from legal and moral domains indicate the effectiveness of the proposed OPO method. Our code and data are released at https://github.com/GAIR-NLP/OPO.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15907?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15907</link>
            <enclosure url="https://arxiv.org/pdf/2312.15907"  type="application/pdf" />
            <author><![CDATA[Chunpu Xu; Steffi Chern; Ethan Chern; Ge Zhang; Zekun Wang; Ruibo Liu; Jing Li; Jie Fu; Pengfei Liu]]></author>
        </item>
        <item>
            <title><![CDATA[Think and Retrieval: A Hypothesis Knowledge Graph Enhanced Medical Large Language Models]]></title>
            <description><![CDATA[<p>We explore how the rise of Large Language Models (LLMs) significantly impacts task performance in the field of Natural Language Processing. We focus on two strategies, Retrieval-Augmented Generation (RAG) and Fine-Tuning (FT), and propose the Hypothesis Knowledge Graph Enhanced (HyKGE) framework, leveraging a knowledge graph to enhance medical LLMs. By integrating LLMs and knowledge graphs, HyKGE demonstrates superior performance in addressing accuracy and interpretability challenges, presenting potential applications in the medical domain. Our evaluations using real-world datasets highlight HyKGE's superiority in providing accurate knowledge with precise confidence, particularly in complex and difficult scenarios. The code will be available until published.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15883?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15883</link>
            <enclosure url="https://arxiv.org/pdf/2312.15883"  type="application/pdf" />
            <author><![CDATA[Xinke Jiang; Ruizhe Zhang; Yongxin Xu; Rihong Qiu; Yue Fang; Zhiyuan Wang; Jinyi Tang; Hongxin Ding; Xu Chu; Junfeng Zhao; Yasha Wang]]></author>
        </item>
        <item>
            <title><![CDATA[KnowledgeNavigator: Leveraging Large Language Models for Enhanced Reasoning over Knowledge Graph]]></title>
            <description><![CDATA[<p>Large language model (LLM) has achieved outstanding performance on various downstream tasks with its powerful natural language understanding and zero-shot capability, but LLM still suffers from knowledge limitation. Especially in scenarios that require long logical chains or complex reasoning, the hallucination and knowledge limitation of LLM limit its performance in question answering (QA). In this paper, we propose a novel framework KnowledgeNavigator to address these challenges by efficiently and accurately retrieving external knowledge from knowledge graph and using it as a key factor to enhance LLM reasoning. Specifically, KnowledgeNavigator first mines and enhances the potential constraints of the given question to guide the reasoning. Then it retrieves and filters external knowledge that supports answering through iterative reasoning on knowledge graph with the guidance of LLM and the question. Finally, KnowledgeNavigator constructs the structured knowledge into effective prompts that are friendly to LLM to help its reasoning. We evaluate KnowledgeNavigator on multiple public KGQA benchmarks, the experiments show the framework has great effectiveness and generalization, outperforming previous knowledge graph enhanced LLM methods and is comparable to the fully supervised models.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15880?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15880</link>
            <enclosure url="https://arxiv.org/pdf/2312.15880"  type="application/pdf" />
            <author><![CDATA[Tiezheng Guo; Qingwen Yang; Chen Wang; Yanyi Liu; Pan Li; Jiawei Tang; Dapeng Li; Yingyou Wen]]></author>
        </item>
        <item>
            <title><![CDATA[Heterogeneous Encoders Scaling In The Transformer For Neural Machine Translation]]></title>
            <description><![CDATA[<p>Although the Transformer is currently the best-performing architecture in the homogeneous configuration (self-attention only) in Neural Machine Translation, many State-of-the-Art models in Natural Language Processing are made of a combination of different Deep Learning approaches. However, these models often focus on combining a couple of techniques only and it is unclear why some methods are chosen over others. In this work, we investigate the effectiveness of integrating an increasing number of heterogeneous methods. Based on a simple combination strategy and performance-driven synergy criteria, we designed the Multi-Encoder Transformer, which consists of up to five diverse encoders. Results showcased that our approach can improve the quality of the translation across a variety of languages and dataset sizes and it is particularly effective in low-resource languages where we observed a maximum increase of 7.16 BLEU compared to the single-encoder model.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15872?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15872</link>
            <enclosure url="https://arxiv.org/pdf/2312.15872"  type="application/pdf" />
            <author><![CDATA[Jia Cheng Hu; Roberto Cavicchioli; Giulia Berardinelli; Alessandro Capotondi]]></author>
        </item>
        <item>
            <title><![CDATA[Medical Report Generation based on Segment-Enhanced Contrastive Representation Learning]]></title>
            <description><![CDATA[<p>Automated radiology report generation has the potential to improve radiology reporting and alleviate the workload of radiologists. However, the medical report generation task poses unique challenges due to the limited availability of medical data and the presence of data bias. To maximize the utility of available data and reduce data bias, we propose MSCL (Medical image Segmentation with Contrastive Learning), a framework that utilizes the Segment Anything Model (SAM) to segment organs, abnormalities, bones, etc., and can pay more attention to the meaningful ROIs in the image to get better visual representations. Then we introduce a supervised contrastive loss that assigns more weight to reports that are semantically similar to the target while training. The design of this loss function aims to mitigate the impact of data bias and encourage the model to capture the essential features of a medical image and generate high-quality reports. Experimental results demonstrate the effectiveness of our proposed model, where we achieve state-of-the-art performance on the IU X-Ray public dataset.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15869?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15869</link>
            <enclosure url="https://arxiv.org/pdf/2312.15869"  type="application/pdf" />
            <author><![CDATA[Ruoqing Zhao; Xi Wang; Hongliang Dai; Pan Gao; Piji Li]]></author>
        </item>
        <item>
            <title><![CDATA[Punctuation Matters! Stealthy Backdoor Attack for Language Models]]></title>
            <description><![CDATA[<p>Recent studies have pointed out that natural language processing (NLP) models are vulnerable to backdoor attacks. A backdoored model produces normal outputs on the clean samples while performing improperly on the texts with triggers that the adversary injects. However, previous studies on textual backdoor attack pay little attention to stealthiness. Moreover, some attack methods even cause grammatical issues or change the semantic meaning of the original texts. Therefore, they can easily be detected by humans or defense systems. In this paper, we propose a novel stealthy backdoor attack method against textual models, which is called \textbf{PuncAttack}. It leverages combinations of punctuation marks as the trigger and chooses proper locations strategically to replace them. Through extensive experiments, we demonstrate that the proposed method can effectively compromise multiple models in various tasks. Meanwhile, we conduct automatic evaluation and human inspection, which indicate the proposed method possesses good performance of stealthiness without bringing grammatical issues and altering the meaning of sentences.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15867?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15867</link>
            <enclosure url="https://arxiv.org/pdf/2312.15867"  type="application/pdf" />
            <author><![CDATA[Xuan Sheng; Zhicheng Li; Zhaoyang Han; Xiangmao Chang; Piji Li]]></author>
        </item>
        <item>
            <title><![CDATA[Knowledge Distillation of LLM for Education]]></title>
            <description><![CDATA[<p>This study proposes a method for distilling the knowledge of fine-tuned Large Language Models (LLMs) into a smaller, more efficient, and accurate neural network, specifically targeting the challenge of deploying these models on resource-constrained devices. Our methodology involves training the smaller student model using the prediction probabilities of the LLM, which serves as a teacher model. This is achieved through a specialized loss function tailored to learn from the LLM's output probabilities, ensuring that the student model closely mimics the teacher's performance. To test this approach, we utilized a large dataset, 7T, containing 6,684 student-written responses to science questions and three other datasets with student-written responses. We also compared performance with original neural network (NN) models to validate the accuracy. Results have shown that the NN and distilled student models have comparable accuracy to the teacher model for the 7T dataset; however, other datasets have shown significantly lower accuracy (28% on average) for NN, though our proposed distilled model is still able to achieve 12\% higher accuracy than NN. Furthermore, the student model size ranges from 0.1M to 0.02M, 100 times smaller in terms of parameters and ten times smaller compared with the original output model size. The significance of this research lies in its potential to make advanced AI technologies accessible in typical educational settings, particularly for automatic scoring.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15842?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15842</link>
            <enclosure url="https://arxiv.org/pdf/2312.15842"  type="application/pdf" />
            <author><![CDATA[Ehsan Latif; Luyang Fang; Ping Ma; Xiaoming Zhai]]></author>
        </item>
        <item>
            <title><![CDATA[SecQA: A Concise Question-Answering Dataset for Evaluating Large Language Models in Computer Security]]></title>
            <description><![CDATA[<p>In this paper, we introduce SecQA, a novel dataset tailored for evaluating the performance of Large Language Models (LLMs) in the domain of computer security. Utilizing multiple-choice questions generated by GPT-4 based on the "Computer Systems Security: Planning for Success" textbook, SecQA aims to assess LLMs' understanding and application of security principles. We detail the structure and intent of SecQA, which includes two versions of increasing complexity, to provide a concise evaluation across various difficulty levels. Additionally, we present an extensive evaluation of prominent LLMs, including GPT-3.5-Turbo, GPT-4, Llama-2, Vicuna, Mistral, and Zephyr models, using both 0-shot and 5-shot learning settings. Our results, encapsulated in the SecQA v1 and v2 datasets, highlight the varying capabilities and limitations of these models in the computer security context. This study not only offers insights into the current state of LLMs in understanding security-related content but also establishes SecQA as a benchmark for future advancements in this critical research area.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15838?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15838</link>
            <enclosure url="https://arxiv.org/pdf/2312.15838"  type="application/pdf" />
            <author><![CDATA[Zefang Liu]]></author>
        </item>
        <item>
            <title><![CDATA[TEILP: Time Prediction over Knowledge Graphs via Logical Reasoning]]></title>
            <description><![CDATA[<p>Conventional embedding-based models approach event time prediction in temporal knowledge graphs (TKGs) as a ranking problem. However, they often fall short in capturing essential temporal relationships such as order and distance. In this paper, we propose TEILP, a logical reasoning framework that naturaly integrates such temporal elements into knowledge graph predictions. We first convert TKGs into a temporal event knowledge graph (TEKG) which has a more explicit representation of time in term of nodes of the graph. The TEKG equips us to develop a differentiable random walk approach to time prediction. Finally, we introduce conditional probability density functions, associated with the logical rules involving the query interval, using which we arrive at the time prediction. We compare TEILP with state-of-the-art methods on five benchmark datasets. We show that our model achieves a significant improvement over baselines while providing interpretable explanations. In particular, we consider several scenarios where training samples are limited, event types are imbalanced, and forecasting the time of future events based on only past events is desired. In all these cases, TEILP outperforms state-of-the-art methods in terms of robustness.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15816?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15816</link>
            <enclosure url="https://arxiv.org/pdf/2312.15816"  type="application/pdf" />
            <author><![CDATA[Siheng Xiong; Yuan Yang; Ali Payani; James C Kerce; Faramarz Fekri]]></author>
        </item>
        <item>
            <title><![CDATA[Compositional Generalization in Spoken Language Understanding]]></title>
            <description><![CDATA[<p>State-of-the-art spoken language understanding (SLU) models have shown tremendous success in benchmark SLU datasets, yet they still fail in many practical scenario due to the lack of model compositionality when trained on limited training data. In this paper, we study two types of compositionality: (a) novel slot combination, and (b) length generalization. We first conduct in-depth analysis, and find that state-of-the-art SLU models often learn spurious slot correlations during training, which leads to poor performance in both compositional cases. To mitigate these limitations, we create the first compositional splits of benchmark SLU datasets and we propose the first compositional SLU model, including compositional loss and paired training that tackle each compositional case respectively. On both benchmark and compositional splits in ATIS and SNIPS, we show that our compositional SLU model significantly outperforms (up to $5\%$ F1 score) state-of-the-art BERT SLU model.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15815?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15815</link>
            <enclosure url="https://arxiv.org/pdf/2312.15815"  type="application/pdf" />
            <author><![CDATA[Avik Ray; Yilin Shen; Hongxia Jin]]></author>
        </item>
        <item>
            <title><![CDATA[AHAM: Adapt, Help, Ask, Model -- Harvesting LLMs for literature mining]]></title>
            <description><![CDATA[<p>In an era marked by a rapid increase in scientific publications, researchers grapple with the challenge of keeping pace with field-specific advances. We present the `AHAM' methodology and a metric that guides the domain-specific \textbf{adapt}ation of the BERTopic topic modeling framework to improve scientific text analysis. By utilizing the LLaMa2 generative language model, we generate topic definitions via one-shot learning by crafting prompts with the \textbf{help} of domain experts to guide the LLM for literature mining by \textbf{asking} it to model the topic names. For inter-topic similarity evaluation, we leverage metrics from language generation and translation processes to assess lexical and semantic similarity of the generated topics. Our system aims to reduce both the ratio of outlier topics to the total number of topics and the similarity between topic definitions. The methodology has been assessed on a newly gathered corpus of scientific papers on literature-based discovery. Through rigorous evaluation by domain experts, AHAM has been validated as effective in uncovering intriguing and novel insights within broad research areas. We explore the impact of domain adaptation of sentence-transformers for the task of topic \textbf{model}ing using two datasets, each specialized to specific scientific domains within arXiv and medarxiv. We evaluate the impact of data size, the niche of adaptation, and the importance of domain adaptation. Our results suggest a strong interaction between domain adaptation and topic modeling precision in terms of outliers and topic definitions.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15784?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15784</link>
            <enclosure url="https://arxiv.org/pdf/2312.15784"  type="application/pdf" />
            <author><![CDATA[Boshko Koloski; Nada Lavrač; Bojan Cestnik; Senja Pollak; Blaž Škrlj; Andrej Kastrin]]></author>
        </item>
        <item>
            <title><![CDATA[Design and Implementation of a Tool for Extracting Uzbek Syllables]]></title>
            <description><![CDATA[<p>The accurate syllabification of words plays a vital role in various Natural Language Processing applications. Syllabification is a versatile linguistic tool with applications in linguistic research, language technology, education, and various fields where understanding and processing language is essential. In this paper, we present a comprehensive approach to syllabification for the Uzbek language, including rule-based techniques and machine learning algorithms. Our rule-based approach utilizes advanced methods for dividing words into syllables, generating hyphenations for line breaks and count of syllables. Additionally, we collected a dataset for evaluating and training using machine learning algorithms comprising word-syllable mappings, hyphenations, and syllable counts to predict syllable counts as well as for the evaluation of the proposed model. Our results demonstrate the effectiveness and efficiency of both approaches in achieving accurate syllabification. The results of our experiments show that both approaches achieved a high level of accuracy, exceeding 99%. This study provides valuable insights and recommendations for future research on syllabification and related areas in not only the Uzbek language itself, but also in other closely-related Turkic languages with low-resource factor.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15779?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15779</link>
            <enclosure url="https://arxiv.org/pdf/2312.15779"  type="application/pdf" />
            <author><![CDATA[Ulugbek Salaev; Elmurod Kuriyozov; Gayrat Matlatipov]]></author>
        </item>
        <item>
            <title><![CDATA[Solving Label Variation in Scientific Information Extraction via Multi-Task Learning]]></title>
            <description><![CDATA[<p>Scientific Information Extraction (ScientificIE) is a critical task that involves the identification of scientific entities and their relationships. The complexity of this task is compounded by the necessity for domain-specific knowledge and the limited availability of annotated data. Two of the most popular datasets for ScientificIE are SemEval-2018 Task-7 and SciERC. They have overlapping samples and differ in their annotation schemes, which leads to conflicts. In this study, we first introduced a novel approach based on multi-task learning to address label variations. We then proposed a soft labeling technique that converts inconsistent labels into probabilistic distributions. The experimental results demonstrated that the proposed method can enhance the model robustness to label noise and improve the end-to-end performance in both ScientificIE tasks. The analysis revealed that label variations can be particularly effective in handling ambiguous instances. Furthermore, the richness of the information captured by label variations can potentially reduce data size requirements. The findings highlight the importance of releasing variation labels and promote future research on other tasks in other domains. Overall, this study demonstrates the effectiveness of multi-task learning and the potential of label variations to enhance the performance of ScientificIE.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15751?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15751</link>
            <enclosure url="https://arxiv.org/pdf/2312.15751"  type="application/pdf" />
            <author><![CDATA[Dong Pham; Xanh Ho; Quang-Thuy Ha; Akiko Aizawa]]></author>
        </item>
        <item>
            <title><![CDATA[PersianLLaMA: Towards Building First Persian Large Language Model]]></title>
            <description><![CDATA[<p>Despite the widespread use of the Persian language by millions globally, limited efforts have been made in natural language processing for this language. The use of large language models as effective tools in various natural language processing tasks typically requires extensive textual data and robust hardware resources. Consequently, the scarcity of Persian textual data and the unavailability of powerful hardware resources have hindered the development of large language models for Persian. This paper introduces the first large Persian language model, named PersianLLaMA, trained on a collection of Persian texts and datasets. This foundational model comes in two versions, with 7 and 13 billion parameters, trained on formal and colloquial Persian texts using two different approaches. PersianLLaMA has been evaluated for natural language generation tasks based on the latest evaluation methods, namely using larger language models, and for natural language understanding tasks based on automated machine metrics. The results indicate that PersianLLaMA significantly outperforms its competitors in both understanding and generating Persian text. PersianLLaMA marks an important step in the development of Persian natural language processing and can be a valuable resource for the Persian-speaking community. This large language model can be used for various natural language processing tasks, especially text generation like chatbots, question-answering, machine translation, and text summarization</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15713?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15713</link>
            <enclosure url="https://arxiv.org/pdf/2312.15713"  type="application/pdf" />
            <author><![CDATA[Mohammad Amin Abbasi; Arash Ghafouri; Mahdi Firouzmandi; Hassan Naderi; Behrouz Minaei Bidgoli]]></author>
        </item>
        <item>
            <title><![CDATA[Alleviating Hallucinations of Large Language Models through Induced Hallucinations]]></title>
            <description><![CDATA[<p>Despite their impressive capabilities, large language models (LLMs) have been observed to generate responses that include inaccurate or fabricated information, a phenomenon commonly known as ``hallucination''. In this work, we propose a simple \textit{Induce-then-Contrast} Decoding (ICD) strategy to alleviate hallucinations. We first construct a factually weak LLM by inducing hallucinations from the original LLMs. Then, we penalize these induced hallucinations during decoding to enhance the factuality of the generated content. Concretely, we determine the final next-token predictions by amplifying the predictions from the original model and downplaying the induced untruthful predictions via contrastive decoding. Experimental results on both discrimination-based and generation-based hallucination evaluation benchmarks, such as TruthfulQA and \textsc{FActScore}, demonstrate that our proposed ICD methods can effectively enhance the factuality of LLMs across various model sizes and families. For example, when equipped with ICD, Llama2-7B-Chat and Mistral-7B-Instruct achieve performance comparable to ChatGPT and GPT4 on TruthfulQA, respectively.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15710?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15710</link>
            <enclosure url="https://arxiv.org/pdf/2312.15710"  type="application/pdf" />
            <author><![CDATA[Yue Zhang; Leyang Cui; Wei Bi; Shuming Shi]]></author>
        </item>
        <item>
            <title><![CDATA[EcomGPT-CT: Continual Pre-training of E-commerce Large Language Models with Semi-structured Data]]></title>
            <description><![CDATA[<p>Large Language Models (LLMs) pre-trained on massive corpora have exhibited remarkable performance on various NLP tasks. However, applying these models to specific domains still poses significant challenges, such as lack of domain knowledge, limited capacity to leverage domain knowledge and inadequate adaptation to domain-specific data formats. Considering the exorbitant cost of training LLMs from scratch and the scarcity of annotated data within particular domains, in this work, we focus on domain-specific continual pre-training of LLMs using E-commerce domain as an exemplar. Specifically, we explore the impact of continual pre-training on LLMs employing unlabeled general and E-commercial corpora. Furthermore, we design a mixing strategy among different data sources to better leverage E-commercial semi-structured data. We construct multiple tasks to assess LLMs' few-shot In-context Learning ability and their zero-shot performance after instruction tuning in E-commerce domain. Experimental results demonstrate the effectiveness of continual pre-training of E-commerce LLMs and the efficacy of our devised data mixing strategy.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15696?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15696</link>
            <enclosure url="https://arxiv.org/pdf/2312.15696"  type="application/pdf" />
            <author><![CDATA[Shirong Ma; Shen Huang; Shulin Huang; Xiaobin Wang; Yangning Li; Hai-Tao Zheng; Pengjun Xie; Fei Huang; Yong Jiang]]></author>
        </item>
        <item>
            <title><![CDATA[What Makes Good Data for Alignment? A Comprehensive Study of Automatic Data Selection in Instruction Tuning]]></title>
            <description><![CDATA[<p>Instruction tuning is a standard technique employed to align large language models to end tasks and user preferences after the initial pretraining phase. Recent research indicates the critical role of data engineering in instruction tuning -- when appropriately selected, only limited data is necessary to achieve superior performance. However, we still lack a principled understanding of what makes good instruction tuning data for alignment, and how we should select data automatically and effectively. In this work, we delve deeply into automatic data selection strategies for alignment. We start with controlled studies to measure data across three dimensions: complexity, quality, and diversity, along which we examine existing methods and introduce novel techniques for enhanced data measurement. Subsequently, we propose a simple strategy to select data samples based on the measurement. We present deita (short for Data-Efficient Instruction Tuning for Alignment), a series of models fine-tuned from LLaMA and Mistral models using data samples automatically selected with our proposed approach. Empirically, deita performs better or on par with the state-of-the-art open-source alignment models with only 6K SFT training data samples -- over 10x less than the data used in the baselines. When further trained with direct preference optimization (DPO), deita-Mistral-7B + DPO trained with 6K SFT and 10K DPO samples achieve 7.55 MT-Bench and 90.06% AlpacaEval scores. We anticipate this work to provide tools on automatic data selection, facilitating data-efficient alignment. We release our models as well as the selected datasets for future researches to effectively align models more efficiently.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15685?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15685</link>
            <enclosure url="https://arxiv.org/pdf/2312.15685"  type="application/pdf" />
            <author><![CDATA[Wei Liu; Weihao Zeng; Keqing He; Yong Jiang; Junxian He]]></author>
        </item>
        <item>
            <title><![CDATA[Conditional Variational Autoencoder for Sign Language Translation with Cross-Modal Alignment]]></title>
            <description><![CDATA[<p>Sign language translation (SLT) aims to convert continuous sign language videos into textual sentences. As a typical multi-modal task, there exists an inherent modality gap between sign language videos and spoken language text, which makes the cross-modal alignment between visual and textual modalities crucial. However, previous studies tend to rely on an intermediate sign gloss representation to help alleviate the cross-modal problem thereby neglecting the alignment across modalities that may lead to compromised results. To address this issue, we propose a novel framework based on Conditional Variational autoencoder for SLT (CV-SLT) that facilitates direct and sufficient cross-modal alignment between sign language videos and spoken language text. Specifically, our CV-SLT consists of two paths with two Kullback-Leibler (KL) divergences to regularize the outputs of the encoder and decoder, respectively. In the prior path, the model solely relies on visual information to predict the target text; whereas in the posterior path, it simultaneously encodes visual information and textual knowledge to reconstruct the target text. The first KL divergence optimizes the conditional variational autoencoder and regularizes the encoder outputs, while the second KL divergence performs a self-distillation from the posterior path to the prior path, ensuring the consistency of decoder outputs. We further enhance the integration of textual information to the posterior path by employing a shared Attention Residual Gaussian Distribution (ARGD), which considers the textual information in the posterior path as a residual component relative to the prior path. Extensive experiments conducted on public datasets (PHOENIX14T and CSL-daily) demonstrate the effectiveness of our framework, achieving new state-of-the-art results while significantly alleviating the cross-modal representation discrepancy.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15645?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15645</link>
            <enclosure url="https://arxiv.org/pdf/2312.15645"  type="application/pdf" />
            <author><![CDATA[Rui Zhao; Liang Zhang; Biao Fu; Cong Hu; Jinsong Su; Yidong Chen]]></author>
        </item>
        <item>
            <title><![CDATA[A Split-and-Privatize Framework for Large Language Model Fine-Tuning]]></title>
            <description><![CDATA[<p>Fine-tuning is a prominent technique to adapt a pre-trained language model to downstream scenarios. In parameter-efficient fine-tuning, only a small subset of modules are trained over the downstream datasets, while leaving the rest of the pre-trained model frozen to save computation resources. In recent years, a popular productization form arises as Model-as-a-Service (MaaS), in which vendors provide abundant pre-trained language models, server resources and core functions, and customers can fine-tune, deploy and invoke their customized model by accessing the one-stop MaaS with their own private dataset. In this paper, we identify the model and data privacy leakage risks in MaaS fine-tuning, and propose a Split-and-Privatize (SAP) framework, which manage to mitigate the privacy issues by adapting the existing split learning architecture. The proposed SAP framework is sufficiently investigated by experiments, and the results indicate that it can enhance the empirical privacy by 62% at the cost of 1% model performance degradation on the Stanford Sentiment Treebank dataset.</p>
  <p><strong style="color:red">访问过快，请稍后重试</strong></p>
]]></description>
            <pubDate>Tue, 26 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.15603?kiwi</guid>
            <link>https://arxiv.org/abs/2312.15603</link>
            <enclosure url="https://arxiv.org/pdf/2312.15603"  type="application/pdf" />
            <author><![CDATA[Xicong Shen; Yang Liu; Huiqi Liu; Jue Hong; Bing Duan; Zirui Huang; Yunlong Mao; Ye Wu; Di Wu]]></author>
        </item>
    </channel>
</rss>

TonyRL · 2023-12-28T17:05:58Z

lib/v2/papers/index.js

+                pubDate,
+                enclosure_url: enclosureUrl,
+                enclosure_type: enclosureUrl ? 'application/pdf' : undefined,
+                doi,


doi always starts with prefix 10.
The actual doi can be found from https://arxiv.org/abs/2312.16171

TonyRL · 2023-12-28T17:20:24Z

lib/v2/papers/index.js

+                    timeout: {
+                        request: 5000,
+                    },


This value is controlled by

RSSHub/lib/utils/got.js

Line 45 in 526d1ea

timeout: config.requestTimeout,

It would be better to not overwrite it.

TonyRL · 2023-12-28T17:21:11Z

lib/v2/papers/index.js

+    items = await Promise.all(
+        items.map(async (item) => {
+            const kimiURL = new URL(item.doi, apiKimiUrl).href;
+
+            const cache = await ctx.cache.get(kimiURL);
+            if (cache) {
+                return Promise.resolve(JSON.parse(cache));
+            }
+
+            try {
+                const { data: detailResponse } = await got(kimiURL, {
+                    timeout: {
+                        request: 5000,
+                    },
+                });
+
+                // Another cache with content by Kiwi Chat.
+
+                item.guid = `${item.guid}?kiwi`;
+                item.description += art(path.join(__dirname, 'templates/description.art'), {
+                    kimi: detailResponse,
+                });
+
+                ctx.cache.set(kimiURL, JSON.stringify(item));
+            } catch (e) {
+                // no-empty
+            }
+
+            return Promise.resolve(item);
+        })
+    );


I think tryGet is fine here.

Suggested change

items = await Promise.all(

items.map(async (item) => {

const kimiURL = new URL(item.doi, apiKimiUrl).href;

const cache = await ctx.cache.get(kimiURL);

if (cache) {

return Promise.resolve(JSON.parse(cache));

}

try {

const { data: detailResponse } = await got(kimiURL, {

timeout: {

request: 5000,

},

});

// Another cache with content by Kiwi Chat.

item.guid = `${item.guid}?kiwi`;

item.description += art(path.join(__dirname, 'templates/description.art'), {

kimi: detailResponse,

});

ctx.cache.set(kimiURL, JSON.stringify(item));

} catch (e) {

// no-empty

}

return Promise.resolve(item);

})

);

items = await Promise.all(

items.map((item) =>

ctx.cache.tryGet(item.guid, async () => {

const kimiURL = new URL(item.doi, apiKimiUrl).href;

try {

const { data: detailResponse } = await got(kimiURL);

// Another cache with content by Kiwi Chat.

item.description += art(path.join(__dirname, 'templates/description.art'), {

kimi: detailResponse,

});

} catch (e) {

// no-empty

}

return item;

})

)

);

Thanks for the review. Since the content from kimi chat is returned in a stream, I'll forgo fetching it and instead place a link to it in the description. 😊

github-actions · 2023-12-29T10:12:34Z

Successfully generated as following:

http://localhost:1200/papers/arxiv/cs.CL - Success ✔️

<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"
>
    <channel>
        <title><![CDATA[Computation and Language | Cool Papers]]></title>
        <link>https://papers.cool/arxiv/cs.CL</link>
        <atom:link href="http://localhost:1200/papers/arxiv/cs.CL" rel="self" type="application/rss+xml" />
        <description><![CDATA[Computation and Language | Cool Papers - Immersive Paper Discovery - Made with love by RSSHub(https://github.com/DIYgod/RSSHub)]]></description>
        <generator>RSSHub</generator>
        <webMaster>i@diygod.me (DIYgod)</webMaster>
        <language>zh-cn</language>
        <lastBuildDate>Fri, 29 Dec 2023 10:12:33 GMT</lastBuildDate>
        <ttl>5</ttl>
        <item>
            <title><![CDATA[Do Androids Know They're Only Dreaming of Electric Sheep?]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17249">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17249">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17249">[Kimi]</a>
  <p>We design probes trained on the internal representations of a transformer language model that are predictive of its hallucinatory behavior on in-context generation tasks. To facilitate this detection, we create a span-annotated dataset of organic and synthetic hallucinations over several tasks. We find that probes trained on the force-decoded states of synthetic hallucinations are generally ecologically invalid in organic hallucination detection. Furthermore, hidden state information about hallucination appears to be task and distribution-dependent. Intrinsic and extrinsic hallucination saliency varies across layers, hidden state types, and tasks; notably, extrinsic hallucinations tend to be more salient in a transformer's internal representations. Outperforming multiple contemporary baselines, we show that probing is a feasible and efficient alternative to language model hallucination evaluation when model states are available.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17249</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17249</link>
            <enclosure url="https://arxiv.org/pdf/2312.17249"  type="application/pdf" />
            <author><![CDATA[Sky CH-Wang; Benjamin Van Durme; Jason Eisner; Chris Kedzie]]></author>
        </item>
        <item>
            <title><![CDATA[Learning to Generate Text in Arbitrary Writing Styles]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17242">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17242">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17242">[Kimi]</a>
  <p>Prior work in style-controlled text generation has focused on tasks such as emulating the style of prolific literary authors, producing formal or informal text, and the degree of toxicity of generated text. Plentiful demonstrations of these styles are available, and as a result modern language models are often able to emulate them, either via prompting or discriminative control. However, in applications such as writing assistants, it is desirable for language models to produce text in an author-specific style on the basis of a small writing sample. We find that instruction-tuned language models can struggle to reproduce author-specific style demonstrated in a prompt. Instead, we propose to guide a language model to generate text in a target style using contrastively-trained representations that capture stylometric features. A central challenge in doing so is that an author's writing is characterized by surprising token choices under a generic language model. To reconcile this tension, we combine generative re-scoring to achieve an author-specific model, with discriminative control to ensure style consistency at the sequence-level. The combination of these approaches is found to be particularly effective at adhering to an author-specific style in a variety of conditions, including unconditional generation and style transfer, and is applicable to any underlying language model without requiring fine-tuning.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17242</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17242</link>
            <enclosure url="https://arxiv.org/pdf/2312.17242"  type="application/pdf" />
            <author><![CDATA[Aleem Khan; Andrew Wang; Sophia Hager; Nicholas Andrews]]></author>
        </item>
        <item>
            <title><![CDATA[Virtual Scientific Companion for Synchrotron Beamlines: A Prototype]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17180">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17180">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17180">[Kimi]</a>
  <p>The extraordinarily high X-ray flux and specialized instrumentation at synchrotron beamlines have enabled versatile in-situ and high throughput studies that are impossible elsewhere. Dexterous and efficient control of experiments are thus crucial for efficient beamline operation. Artificial intelligence and machine learning methods are constantly being developed to enhance facility performance, but the full potential of these developments can only be reached with efficient human-computer-interaction. Natural language is the most intuitive and efficient way for humans to communicate. However, the low credibility and reproducibility of existing large language models and tools demand extensive development to be made for robust and reliable performance for scientific purposes. In this work, we introduce the prototype of virtual scientific companion (VISION) and demonstrate that it is possible to control basic beamline operations through natural language with open-source language model and the limited computational resources at beamline. The human-AI nature of VISION leverages existing automation systems and data framework at synchrotron beamlines.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17180</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17180</link>
            <enclosure url="https://arxiv.org/pdf/2312.17180"  type="application/pdf" />
            <author><![CDATA[Daniel Potemkin; Carlos Soto; Ruipeng Li; Kevin Yager; Esther Tsai]]></author>
        </item>
        <item>
            <title><![CDATA[Large Language Model for Causal Decision Making]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17122">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17122">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17122">[Kimi]</a>
  <p>Large Language Models (LLMs) have shown their success in language understanding and reasoning on general topics. However, their capability to inference based on user-specified structured data and knowledge in corpus-rare concepts like causal decision-making is still limited. In this work, we explore the possibility of fine-tuning an open-sourced LLM into LLM4Causal, which can identify the causal task, execute a corresponding function, and interpret its numerical results based on users' queries and the provided dataset. Meanwhile, we propose a data generation process for more controllable GPT prompting and present two instruction-tuning datasets: (1) Causal-Retrieval-Bench for causal problem identification and input parameter extraction for causal function calling and (2) Causal-Interpret-Bench for in-context causal interpretation. With three case studies, we showed that LLM4Causal can deliver end-to-end solutions for causal problems and provide easy-to-understand answers. Numerical studies also reveal that it has a remarkable ability to identify the correct causal task given a query.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17122</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17122</link>
            <enclosure url="https://arxiv.org/pdf/2312.17122"  type="application/pdf" />
            <author><![CDATA[Haitao Jiang; Lin Ge; Yuhe Gao; Jianian Wang; Rui Song]]></author>
        </item>
        <item>
            <title><![CDATA[Generative AI for Math: Part I -- MathPile: A Billion-Token-Scale Pretraining Corpus for Math]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17120">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17120">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17120">[Kimi]</a>
  <p>High-quality, large-scale corpora are the cornerstone of building foundation models. In this work, we introduce \textsc{MathPile}, a diverse and high-quality math-centric corpus comprising about 9.5 billion tokens. Throughout its creation, we adhered to the principle of ``\emph{less is more}'', firmly believing in the supremacy of data quality over quantity, even in the pre-training phase. Our meticulous data collection and processing efforts included a complex suite of preprocessing, prefiltering, language identification, cleaning, filtering, and deduplication, ensuring the high quality of our corpus. Furthermore, we performed data contamination detection on downstream benchmark test sets to eliminate duplicates. We hope our \textsc{MathPile} can help to enhance the mathematical reasoning abilities of language models. We plan to open-source different versions of \mathpile with the scripts used for processing, to facilitate future developments in this field.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17120</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17120</link>
            <enclosure url="https://arxiv.org/pdf/2312.17120"  type="application/pdf" />
            <author><![CDATA[Zengzhi Wang; Rui Xia; Pengfei Liu]]></author>
        </item>
        <item>
            <title><![CDATA[How Far Are We from Believable AI Agents? A Framework for Evaluating the Believability of Human Behavior Simulation]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17115">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17115">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17115">[Kimi]</a>
  <p>Human behavior simulation of AI agents necessitates the agents to possess a quality of believability, which is crucial as it facilitates users in establishing trust toward the agents and streamlines the fulfillment of the agents' goal. While recent advancements in Large Language Model (LLM) based agents have improved human behavior simulation, challenges inherent to LLMs (e.g., long context modeling) can undermine their believability. Consequently, evaluating AI agent believability becomes imperative. Unfortunately, prior research often neglects the negative impacts of LLM deficiencies. To address these gaps, we introduce two metrics for assessing LLM-based agent believability: consistency, and robustness, together with a benchmark, SimulateBench, with which, we evaluate the consistency and robustness of agents implemented with popular LLMs. We find that agents (i) struggle to accurately depict character information when presented with lengthy profile inputs; (ii) exhibit vulnerability to profile perturbations; and (iii) are significantly affected by certain key factors that impact their overall believability. Code and SimulateBench are public at https://github.com/GAIR-NLP/GPTMan.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17115</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17115</link>
            <enclosure url="https://arxiv.org/pdf/2312.17115"  type="application/pdf" />
            <author><![CDATA[Yang Xiao; Yi Cheng; Jinlan Fu; Jiashuo Wang; Wenjie Li; Pengfei Liu]]></author>
        </item>
        <item>
            <title><![CDATA[Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17080">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17080">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17080">[Kimi]</a>
  <p>In this work, we introduce a novel evaluation paradigm for Large Language Models, one that challenges them to engage in meta-reasoning. This approach addresses critical shortcomings in existing math problem-solving benchmarks, traditionally used to evaluate the cognitive capabilities of agents. Our paradigm shifts the focus from result-oriented assessments, which often overlook the reasoning process, to a more holistic evaluation that effectively differentiates the cognitive capabilities among models. For example, in our benchmark, GPT-4 demonstrates a performance ten times more accurate than GPT3-5. The significance of this new paradigm lies in its ability to reveal potential cognitive deficiencies in LLMs that current benchmarks, such as GSM8K, fail to uncover due to their saturation and lack of effective differentiation among varying reasoning abilities. Our comprehensive analysis includes several state-of-the-art math models from both open-source and closed-source communities, uncovering fundamental deficiencies in their training and evaluation approaches. This paper not only advocates for a paradigm shift in the assessment of LLMs but also contributes to the ongoing discourse on the trajectory towards Artificial General Intelligence (AGI). By promoting the adoption of meta-reasoning evaluation methods similar to ours, we aim to facilitate a more accurate assessment of the true cognitive abilities of LLMs.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17080</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17080</link>
            <enclosure url="https://arxiv.org/pdf/2312.17080"  type="application/pdf" />
            <author><![CDATA[Zhongshen Zeng; Pengguang Chen; Haiyun Jiang; Jiaya Jia]]></author>
        </item>
        <item>
            <title><![CDATA[Improving In-context Learning via Bidirectional Alignment]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17055">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17055">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17055">[Kimi]</a>
  <p>Large language models (LLMs) have shown impressive few-shot generalization on many tasks via in-context learning (ICL). Despite their success in showing such emergent abilities, the scale and complexity of larger models also lead to unprecedentedly high computational demands and deployment challenges. In reaction, researchers explore transferring the powerful capabilities of larger models to more efficient and compact models by typically aligning the output of smaller models with that of larger models. Existing methods either train smaller models on the generated outputs of larger models or to imitate their token-level probability distributions. However, these distillation methods pay little to no attention to the input part, which also plays a crucial role in ICL. Based on the finding that the performance of ICL is highly sensitive to the selection of demonstration examples, we propose Bidirectional Alignment (BiAlign) to fully leverage the models' preferences for ICL examples to improve the ICL abilities of smaller models. Specifically, we introduce the alignment of input preferences between smaller and larger models by incorporating a novel ranking loss, in addition to aligning the token-level output distribution. With extensive experiments and analysis, we demonstrate that BiAlign can consistently outperform existing baselines on a variety of tasks including language understanding, reasoning, and coding.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17055</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17055</link>
            <enclosure url="https://arxiv.org/pdf/2312.17055"  type="application/pdf" />
            <author><![CDATA[Chengwei Qin; Wenhan Xia; Fangkai Jiao; Shafiq Joty]]></author>
        </item>
        <item>
            <title><![CDATA[Length Extrapolation of Transformers: A Survey from the Perspective of Position Encoding]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17044">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17044">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17044">[Kimi]</a>
  <p>Transformer has taken the natural language processing (NLP) field by storm since birth, owing to its superior ability to model complex dependencies in sequences. Despite the great success of pretrained language models (PLMs) based on Transformer across almost all NLP tasks, they all suffer from a preset length limit and thus can hardly extend this success to longer sequences beyond seen data, namely the length extrapolation problem. Length extrapolation has aroused great interest among researchers, as it is the core feature of human language capacity. To enhance length extrapolation of Transformers, a plethora of methods have been proposed, mostly focusing on extrapolatable position encodings. In this article, we provide an organized and systematical review of these research efforts in a unified notation from a position encoding perspective, aiming to enable the reader to gain a deep understanding of existing methods and provide stimuli for future research.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17044</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17044</link>
            <enclosure url="https://arxiv.org/pdf/2312.17044"  type="application/pdf" />
            <author><![CDATA[Liang Zhao; Xiaocheng Feng; Xiachong Feng; Bin Qin; Ting Liu]]></author>
        </item>
        <item>
            <title><![CDATA[Experiential Co-Learning of Software-Developing Agents]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17025">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17025">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17025">[Kimi]</a>
  <p>Recent advancements in large language models (LLMs) have brought significant changes to various dimains, especially through LLM-driven autonomous agents. These agents are now capable of collaborating seamlessly, splitting tasks and enhancing accuracy, thus minimizing the need for human involvement. However, these agents often approach a diverse range of tasks in isolation, without benefiting from past experiences. This isolation can lead to repeated mistakes and inefficient trials in task solving. To this end, this paper introduces Experiential Co-Learning, a novel framework in which instructor and assistant agents gather shortcut-oriented experiences from their historical trajectories and use these past experiences for mutual reasoning. This paradigm, enriched with previous experiences, equips agents to more effectively address unseen tasks.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17025</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17025</link>
            <enclosure url="https://arxiv.org/pdf/2312.17025"  type="application/pdf" />
            <author><![CDATA[Chen Qian; Yufan Dang; Jiahao Li; Wei Liu; Weize Chen; Cheng Yang; Zhiyuan Liu; Maosong Sun]]></author>
        </item>
        <item>
            <title><![CDATA[Few-shot learning for automated content analysis: Efficient coding of arguments and claims in the debate on arms deliveries to Ukraine]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16975">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16975">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16975">[Kimi]</a>
  <p>Pre-trained language models (PLM) based on transformer neural networks developed in the field of natural language processing (NLP) offer great opportunities to improve automatic content analysis in communication science, especially for the coding of complex semantic categories in large datasets via supervised machine learning. However, three characteristics so far impeded the widespread adoption of the methods in the applying disciplines: the dominance of English language models in NLP research, the necessary computing resources, and the effort required to produce training data to fine-tune PLMs. In this study, we address these challenges by using a multilingual transformer model in combination with the adapter extension to transformers, and few-shot learning methods. We test our approach on a realistic use case from communication science to automatically detect claims and arguments together with their stance in the German news debate on arms deliveries to Ukraine. In three experiments, we evaluate (1) data preprocessing strategies and model variants for this task, (2) the performance of different few-shot learning methods, and (3) how well the best setup performs on varying training set sizes in terms of validity, reliability, replicability and reproducibility of the results. We find that our proposed combination of transformer adapters with pattern exploiting training provides a parameter-efficient and easily shareable alternative to fully fine-tuning PLMs. It performs on par in terms of validity, while overall, provides better properties for application in communication studies. The results also show that pre-fine-tuning for a task on a near-domain dataset leads to substantial improvement, in particular in the few-shot setting. Further, the results indicate that it is useful to bias the dataset away from the viewpoints of specific prominent individuals.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16975</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16975</link>
            <enclosure url="https://arxiv.org/pdf/2312.16975"  type="application/pdf" />
            <author><![CDATA[Jonas Rieger; Kostiantyn Yanchenko; Mattes Ruckdeschel; Gerret von Nordheim; Katharina Kleinen-von Königslöw; Gregor Wiedemann]]></author>
        </item>
        <item>
            <title><![CDATA[Unified Lattice Graph Fusion for Chinese Named Entity Recognition]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16917">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16917">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16917">[Kimi]</a>
  <p>Integrating lexicon into character-level sequence has been proven effective to leverage word boundary and semantic information in Chinese named entity recognition (NER). However, prior approaches usually utilize feature weighting and position coupling to integrate word information, but ignore the semantic and contextual correspondence between the fine-grained semantic units in the character-word space. To solve this issue, we propose a Unified Lattice Graph Fusion (ULGF) approach for Chinese NER. ULGF can explicitly capture various semantic and boundary relations across different semantic units with the adjacency matrix by converting the lattice structure into a unified graph. We stack multiple graph-based intra-source self-attention and inter-source cross-gating fusion layers that iteratively carry out semantic interactions to learn node representations. To alleviate the over-reliance on word information, we further propose to leverage lexicon entity classification as an auxiliary task. Experiments on four Chinese NER benchmark datasets demonstrate the superiority of our ULGF approach.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16917</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16917</link>
            <enclosure url="https://arxiv.org/pdf/2312.16917"  type="application/pdf" />
            <author><![CDATA[Dixiang Zhang; Junyu Lu; Pingjian Zhang]]></author>
        </item>
        <item>
            <title><![CDATA[Spike No More: Stabilizing the Pre-training of Large Language Models]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16903">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16903">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16903">[Kimi]</a>
  <p>The loss spike often occurs during pre-training of a large language model. The spikes degrade the performance of a large language model, and sometimes ruin the pre-training. Since the pre-training needs a vast computational budget, we should avoid such spikes. To investigate a cause of loss spikes, we focus on gradients of internal layers in this study. Through theoretical analyses, we introduce two causes of the exploding gradients, and provide requirements to prevent the explosion. In addition, we introduce the combination of the initialization method and a simple modification to embeddings as a method to satisfy the requirements. We conduct various experiments to verify our theoretical analyses empirically. Experimental results indicate that the combination is effective in preventing spikes during pre-training.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16903</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16903</link>
            <enclosure url="https://arxiv.org/pdf/2312.16903"  type="application/pdf" />
            <author><![CDATA[Sho Takase; Shun Kiyono; Sosuke Kobayashi; Jun Suzuki]]></author>
        </item>
        <item>
            <title><![CDATA[BBScore: A Brownian Bridge Based Metric for Assessing Text Coherence]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16893">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16893">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16893">[Kimi]</a>
  <p>Measuring the coherence of text is a vital aspect of evaluating the quality of written content. Recent advancements in neural coherence modeling have demonstrated their efficacy in capturing entity coreference and discourse relations, thereby enhancing coherence evaluation. However, many existing methods heavily depend on static embeddings or focus narrowly on nearby context, constraining their capacity to measure the overarching coherence of long texts. In this paper, we posit that coherent texts inherently manifest a sequential and cohesive interplay among sentences, effectively conveying the central theme, purpose, or standpoint. To explore this abstract relationship, we introduce the "BBScore," a novel reference-free metric grounded in Brownian bridge theory for assessing text coherence. Our findings showcase that when synergized with a simple additional classification component, this metric attains a performance level comparable to state-of-the-art techniques on standard artificial discrimination tasks. We also establish in downstream tasks that this metric effectively differentiates between human-written documents and text generated by large language models under a specific domain. Furthermore, we illustrate the efficacy of this approach in detecting written styles attributed to diverse large language models, underscoring its potential for generalizability. In summary, we present a novel Brownian bridge coherence metric capable of measuring both local and global text coherence, while circumventing the need for end-to-end model training. This flexibility allows for its application in various downstream tasks.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16893</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16893</link>
            <enclosure url="https://arxiv.org/pdf/2312.16893"  type="application/pdf" />
            <author><![CDATA[Zhecheng Sheng; Tianhao Zhang; Chen Jiang; Dongyeop Kang]]></author>
        </item>
        <item>
            <title><![CDATA[OmniDialog: An Omnipotent Pre-training Model for Task-Oriented Dialogue System]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16864">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16864">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16864">[Kimi]</a>
  <p>Pre-trained conversation models (PCMs) have demonstrated remarkable results in task-oriented dialogue (TOD) systems. Many PCMs focus predominantly on dialogue management tasks like dialogue state tracking, dialogue generation tasks like response generation, or both. However, the existing PCMs seldom consider dialogue comprehension tasks, such as dialogue question answering and summarization tasks. These tasks allow PCMs to glean dialogue context from various angles. This observation naturally raises the question: Can the performance of downstream dialogue tasks be enhanced if a PCM is pre-trained on dialogue management, generation, and comprehension tasks? To investigate this, we proposed an Omnipotent Dialogue pre-training model (OmniDialog). It unifies these three dialogue tasks into a monolithic framework by multi-task learning, fostering inter-task communication. The pre-training corpus of OmniDialog spans $\mathbf{7}$ dialogue-focused tasks, drawing from $\mathbf{15}$ datasets and encompassing over $\mathbf{3.2}$ million dialogue utterances. To our knowledge, OmniDialog is a pioneering PCM pre-trained across dialogue management, generation, and comprehension domains. We evaluated its performance across four tasks: dialogue summarization, end-to-end dialogue modeling, dialogue state tracking, and intent classification. The results underscore its efficacy in domain transfer learning, low-resource, and full-dataset scenarios. Furthermore, to glean a nuanced understanding of OmniDialog's strengths and potential pitfalls, we designed a fine-grained analysis framework for dialogue-centric tasks. Experimental results show that the OmniDialog is good at hard samples, such as long dialogues and lengthy responses.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16864</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16864</link>
            <enclosure url="https://arxiv.org/pdf/2312.16864"  type="application/pdf" />
            <author><![CDATA[Mingtao Yang; See-Kiong Ng; Jinlan Fu]]></author>
        </item>
        <item>
            <title><![CDATA[Evaluating the Performance of Large Language Models for Spanish Language in Undergraduate Admissions Exams]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16845">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16845">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16845">[Kimi]</a>
  <p>This study evaluates the performance of large language models, specifically GPT-3.5 and BARD (supported by Gemini Pro model), in undergraduate admissions exams proposed by the National Polytechnic Institute in Mexico. The exams cover Engineering/Mathematical and Physical Sciences, Biological and Medical Sciences, and Social and Administrative Sciences. Both models demonstrated proficiency, exceeding the minimum acceptance scores for respective academic programs to up to 75% for some academic programs. GPT-3.5 outperformed BARD in Mathematics and Physics, while BARD performed better in History and questions related to factual information. Overall, GPT-3.5 marginally surpassed BARD with scores of 60.94% and 60.42%, respectively.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16845</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16845</link>
            <enclosure url="https://arxiv.org/pdf/2312.16845"  type="application/pdf" />
            <author><![CDATA[Sabino Miranda; Obdulia Pichardo-Lagunas; Bella Martínez-Seis; Pierre Baldi]]></author>
        </item>
        <item>
            <title><![CDATA[Hiding in Plain Sight: Towards the Science of Linguistic Steganography]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16840">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16840">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16840">[Kimi]</a>
  <p>Covert communication (also known as steganography) is the practice of concealing a secret inside an innocuous-looking public object (cover) so that the modified public object (covert code) makes sense to everyone but only someone who knows the code can extract the secret (message). Linguistic steganography is the practice of encoding a secret message in natural language text such as spoken conversation or short public communications such as tweets.. While ad hoc methods for covert communications in specific domains exist ( JPEG images, Chinese poetry, etc), there is no general model for linguistic steganography specifically. We present a novel mathematical formalism for creating linguistic steganographic codes, with three parameters: Decodability (probability that the receiver of the coded message will decode the cover correctly), density (frequency of code words in a cover code), and detectability (probability that an attacker can tell the difference between an untampered cover compared to its steganized version). Verbal or linguistic steganography is most challenging because of its lack of artifacts to hide the secret message in. We detail a practical construction in Python of a steganographic code for Tweets using inserted words to encode hidden digits while using n-gram frequency distortion as the measure of detectability of the insertions. Using the publicly accessible Stanford Sentiment Analysis dataset we implemented the tweet steganization scheme -- a codeword (an existing word in the data set) inserted in random positions in random existing tweets to find the tweet that has the least possible n-gram distortion. We argue that this approximates KL distance in a localized manner at low cost and thus we get a linguistic steganography scheme that is both formal and practical and permits a tradeoff between codeword density and detectability of the covert message.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16840</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16840</link>
            <enclosure url="https://arxiv.org/pdf/2312.16840"  type="application/pdf" />
            <author><![CDATA[Leela Raj-Sankar; S. Raj Rajagopalan]]></author>
        </item>
        <item>
            <title><![CDATA[Adversarial Representation with Intra-Modal and Inter-Modal Graph Contrastive Learning for Multimodal Emotion Recognition]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16778">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16778">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16778">[Kimi]</a>
  <p>With the release of increasing open-source emotion recognition datasets on social media platforms and the rapid development of computing resources, multimodal emotion recognition tasks (MER) have begun to receive widespread research attention. The MER task extracts and fuses complementary semantic information from different modalities, which can classify the speaker's emotions. However, the existing feature fusion methods have usually mapped the features of different modalities into the same feature space for information fusion, which can not eliminate the heterogeneity between different modalities. Therefore, it is challenging to make the subsequent emotion class boundary learning. To tackle the above problems, we have proposed a novel Adversarial Representation with Intra-Modal and Inter-Modal Graph Contrastive for Multimodal Emotion Recognition (AR-IIGCN) method. Firstly, we input video, audio, and text features into a multi-layer perceptron (MLP) to map them into separate feature spaces. Secondly, we build a generator and a discriminator for the three modal features through adversarial representation, which can achieve information interaction between modalities and eliminate heterogeneity among modalities. Thirdly, we introduce contrastive graph representation learning to capture intra-modal and inter-modal complementary semantic information and learn intra-class and inter-class boundary information of emotion categories. Specifically, we construct a graph structure for three modal features and perform contrastive representation learning on nodes with different emotions in the same modality and the same emotion in different modalities, which can improve the feature representation ability of nodes. Extensive experimental works show that the ARL-IIGCN method can significantly improve emotion recognition accuracy on IEMOCAP and MELD datasets.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16778</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16778</link>
            <enclosure url="https://arxiv.org/pdf/2312.16778"  type="application/pdf" />
            <author><![CDATA[Yuntao Shou; Tao Meng; Wei Ai; Keqin Li]]></author>
        </item>
        <item>
            <title><![CDATA[Graph Neural Networks for Antisocial Behavior Detection on Twitter]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16755">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16755">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16755">[Kimi]</a>
  <p>Social media resurgence of antisocial behavior has exerted a downward spiral on stereotypical beliefs, and hateful comments towards individuals and social groups, as well as false or distorted news. The advances in graph neural networks employed on massive quantities of graph-structured data raise high hopes for the future of mediating communication on social media platforms. An approach based on graph convolutional data was employed to better capture the dependencies between the heterogeneous types of data. Utilizing past and present experiences on the topic, we proposed and evaluated a graph-based approach for antisocial behavior detection, with general applicability that is both language- and context-independent. In this research, we carried out an experimental validation of our graph-based approach on several PAN datasets provided as part of their shared tasks, that enable the discussion of the results obtained by the proposed solution.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16755</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16755</link>
            <enclosure url="https://arxiv.org/pdf/2312.16755"  type="application/pdf" />
            <author><![CDATA[Martina Toshevska; Slobodan Kalajdziski; Sonja Gievska]]></author>
        </item>
        <item>
            <title><![CDATA[A Reversible Perspective on Petri Nets and Event Structures]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16714">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16714">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16714">[Kimi]</a>
  <p>Event structures have emerged as a foundational model for concurrent computation, explaining computational processes by outlining the events and the relationships that dictate their execution. They play a pivotal role in the study of key aspects of concurrent computation models, such as causality and independence, and have found applications across a broad range of languages and models, spanning realms like persistence, probabilities, and quantum computing. Recently, event structures have been extended to address reversibility, where computational processes can undo previous computations. In this context, reversible event structures provide abstract representations of processes capable of both forward and backward steps in a computation. Since their introduction, event structures have played a crucial role in bridging operational models, traditionally exemplified by Petri nets and process calculi, with denotational ones, i.e., algebraic domains. In this context, we revisit the standard connection between Petri nets and event structures under the lenses of reversibility. Specifically, we introduce a subset of contextual Petri nets, dubbed reversible causal nets, that precisely correspond to reversible prime event structures. The distinctive feature of reversible causal nets lies in deriving causality from inhibitor arcs, departing from the conventional dependence on the overlap between the post and preset of transitions. In this way, we are able to operationally explain the full model of reversible prime event structures.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16714</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16714</link>
            <enclosure url="https://arxiv.org/pdf/2312.16714"  type="application/pdf" />
            <author><![CDATA[Hernán Melgratti; Claudio Antares Mezzina; G. Michele Pinna]]></author>
        </item>
        <item>
            <title><![CDATA[Rethinking Tabular Data Understanding with Large Language Models]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16702">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16702">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16702">[Kimi]</a>
  <p>Large Language Models (LLMs) have shown to be capable of various tasks, yet their capability in interpreting and reasoning over tabular data remains an underexplored area. In this context, this study investigates from three core perspectives: the robustness of LLMs to structural perturbations in tables, the comparative analysis of textual and symbolic reasoning on tables, and the potential of boosting model performance through the aggregation of multiple reasoning pathways. We discover that structural variance of tables presenting the same content reveals a notable performance decline, particularly in symbolic reasoning tasks. This prompts the proposal of a method for table structure normalization. Moreover, textual reasoning slightly edges out symbolic reasoning, and a detailed error analysis reveals that each exhibits different strengths depending on the specific tasks. Notably, the aggregation of textual and symbolic reasoning pathways, bolstered by a mix self-consistency mechanism, resulted in achieving SOTA performance, with an accuracy of 73.6% on WIKITABLEQUESTIONS, representing a substantial advancement over previous existing table processing paradigms of LLMs.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16702</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16702</link>
            <enclosure url="https://arxiv.org/pdf/2312.16702"  type="application/pdf" />
            <author><![CDATA[Tianyang Liu; Fei Wang; Muhao Chen]]></author>
        </item>
        <item>
            <title><![CDATA[Some things are more CRINGE than others: Preference Optimization with the Pairwise Cringe Loss]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16682">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16682">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16682">[Kimi]</a>
  <p>Practitioners commonly align large language models using pairwise preferences, i.e., given labels of the type response A is preferred to response B for a given input. Perhaps less commonly, methods have also been developed for binary feedback, i.e. training models given labels of type response A is good or bad. We show how an existing performant binary feedback method, the Cringe Loss (Adolphs et al., 2022), can be generalized to the pairwise preference setting using a simple soft margin extension. Pairwise Cringe Loss is straightforward to implement and efficient to train, and we find it outperforms state-of-the-art preference optimization algorithms such as PPO and DPO on the AlpacaFarm benchmark.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16682</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16682</link>
            <enclosure url="https://arxiv.org/pdf/2312.16682"  type="application/pdf" />
            <author><![CDATA[Jing Xu; Andrew Lee; Sainbayar Sukhbaatar; Jason Weston]]></author>
        </item>
        <item>
            <title><![CDATA[A Large Language Model-based Computational Approach to Improve Identity-Related Write-Ups]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16659">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16659">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16659">[Kimi]</a>
  <p>Creating written products is essential to modern life, including writings about one's identity and personal experiences. However, writing is often a difficult activity that requires extensive effort to frame the central ideas, the pursued approach to communicate the central ideas, e.g., using analogies, metaphors, or other possible means, the needed presentation structure, and the actual verbal expression. Large Language Models, a recently emerged approach in Machine Learning, can offer a significant help in reducing the effort and improving the quality of written products. This paper proposes a new computational approach to explore prompts that given as inputs to a Large Language Models can generate cues to improve the considered written products. Two case studies on improving write-ups, one based on an analogy and one on a metaphor, are also presented in the paper.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16659</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16659</link>
            <enclosure url="https://arxiv.org/pdf/2312.16659"  type="application/pdf" />
            <author><![CDATA[Alex Doboli]]></author>
        </item>
        <item>
            <title><![CDATA[Make BERT-based Chinese Spelling Check Model Enhanced by Layerwise Attention and Gaussian Mixture Model]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16623">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16623">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16623">[Kimi]</a>
  <p>BERT-based models have shown a remarkable ability in the Chinese Spelling Check (CSC) task recently. However, traditional BERT-based methods still suffer from two limitations. First, although previous works have identified that explicit prior knowledge like Part-Of-Speech (POS) tagging can benefit in the CSC task, they neglected the fact that spelling errors inherent in CSC data can lead to incorrect tags and therefore mislead models. Additionally, they ignored the correlation between the implicit hierarchical information encoded by BERT's intermediate layers and different linguistic phenomena. This results in sub-optimal accuracy. To alleviate the above two issues, we design a heterogeneous knowledge-infused framework to strengthen BERT-based CSC models. To incorporate explicit POS knowledge, we utilize an auxiliary task strategy driven by Gaussian mixture model. Meanwhile, to incorporate implicit hierarchical linguistic knowledge within the encoder, we propose a novel form of n-gram-based layerwise self-attention to generate a multilayer representation. Experimental results show that our proposed framework yields a stable performance boost over four strong baseline models and outperforms the previous state-of-the-art methods on two datasets.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16623</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16623</link>
            <enclosure url="https://arxiv.org/pdf/2312.16623"  type="application/pdf" />
            <author><![CDATA[Yongchang Cao; Liang He; Zhen Wu; Xinyu Dai]]></author>
        </item>
        <item>
            <title><![CDATA[Relationship between auditory and semantic entrainment using Deep Neural Networks (DNN)]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16599">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16599">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16599">[Kimi]</a>
  <p>The tendency of people to engage in similar, matching, or synchronized behaviour when interacting is known as entrainment. Many studies examined linguistic (syntactic and lexical structures) and paralinguistic (pitch, intensity) entrainment, but less attention was given to finding the relationship between them. In this study, we utilized state-of-the-art DNN embeddings such as BERT and TRIpLet Loss network (TRILL) vectors to extract features for measuring semantic and auditory similarities of turns within dialogues in two comparable spoken corpora of two different languages. We found people's tendency to entrain on semantic features more when compared to auditory features. Additionally, we found that entrainment in semantic and auditory linguistic features are positively correlated. The findings of this study might assist in implementing the mechanism of entrainment in human-machine interaction (HMI).</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16599</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16599</link>
            <enclosure url="https://arxiv.org/pdf/2312.16599"  type="application/pdf" />
            <author><![CDATA[Jay Kejriwal; Štefan Beňuš]]></author>
        </item>
        <item>
            <title><![CDATA[A proposed new metric for the conceptual diversity of a text]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16548">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16548">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16548">[Kimi]</a>
  <p>A word may contain one or more hidden concepts. While the "animal" word evokes many images in our minds and encapsulates many concepts (birds, dogs, cats, crocodiles, etc.), the `parrot' word evokes a single image (a colored bird with a short, hooked beak and the ability to mimic sounds). In spoken or written texts, we use some words in a general sense and some in a detailed way to point to a specific object. Until now, a text's conceptual diversity value cannot be determined using a standard and precise technique. This research contributes to the natural language processing field of AI by offering a standardized method and a generic metric for evaluating and comparing concept diversity in different texts and domains. It also contributes to the field of semantic research of languages. If we give examples for the diversity score of two sentences, "He discovered an unknown entity." has a high conceptual diversity score (16.6801), and "The endoplasmic reticulum forms a series of flattened sacs within the cytoplasm of eukaryotic cells." sentence has a low conceptual diversity score which is 3.9068.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16548</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16548</link>
            <enclosure url="https://arxiv.org/pdf/2312.16548"  type="application/pdf" />
            <author><![CDATA[İlknur Dönmez Phd; Mehmet Haklıdır Phd]]></author>
        </item>
        <item>
            <title><![CDATA[S2M: Converting Single-Turn to Multi-Turn Datasets for Conversational Question Answering]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16511">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16511">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16511">[Kimi]</a>
  <p>Supplying data augmentation to conversational question answering (CQA) can effectively improve model performance. However, there is less improvement from single-turn datasets in CQA due to the distribution gap between single-turn and multi-turn datasets. On the other hand, while numerous single-turn datasets are available, we have not utilized them effectively. To solve this problem, we propose a novel method to convert single-turn datasets to multi-turn datasets. The proposed method consists of three parts, namely, a QA pair Generator, a QA pair Reassembler, and a question Rewriter. Given a sample consisting of context and single-turn QA pairs, the Generator obtains candidate QA pairs and a knowledge graph based on the context. The Reassembler utilizes the knowledge graph to get sequential QA pairs, and the Rewriter rewrites questions from a conversational perspective to obtain a multi-turn dataset S2M. Our experiments show that our method can synthesize effective training resources for CQA. Notably, S2M ranks 1st place on the QuAC leaderboard at the time of submission (Aug 24th, 2022).</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16511</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16511</link>
            <enclosure url="https://arxiv.org/pdf/2312.16511"  type="application/pdf" />
            <author><![CDATA[Baokui Li; Sen Zhang; Wangshu Zhang; Yicheng Chen; Changlin Yang; Sen Hu; Teng Xu; Siye liu; Jiwei Li]]></author>
        </item>
        <item>
            <title><![CDATA[Understanding News Creation Intents: Frame, Dataset, and Method]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16490">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16490">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16490">[Kimi]</a>
  <p>As the disruptive changes in the media economy and the proliferation of alternative news media outlets, news intent has progressively deviated from ethical standards that serve the public interest. News intent refers to the purpose or intention behind the creation of a news article. While the significance of research on news intent has been widely acknowledged, the absence of a systematic news intent understanding framework hinders further exploration of news intent and its downstream applications. To bridge this gap, we propose News INTent (NINT) frame, the first component-aware formalism for understanding the news creation intent based on research in philosophy, psychology, and cognitive science. Within this frame, we define the news intent identification task and provide a benchmark dataset with fine-grained labels along with an efficient benchmark method. Experiments demonstrate that NINT is beneficial in both the intent identification task and downstream tasks that demand a profound understanding of news. This work marks a foundational step towards a more systematic exploration of news creation intents.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16490</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16490</link>
            <enclosure url="https://arxiv.org/pdf/2312.16490"  type="application/pdf" />
            <author><![CDATA[Zhengjia Wang; Danding Wang; Qiang Sheng; Juan Cao; Silong Su; Yifan Sun; Beizhe Hu; Siyuan Ma]]></author>
        </item>
        <item>
            <title><![CDATA[Source Code is a Graph, Not a Sequence: A Cross-Lingual Perspective on Code Clone Detection]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16488">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16488">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16488">[Kimi]</a>
  <p>Source code clone detection is the task of finding code fragments that have the same or similar functionality, but may differ in syntax or structure. This task is important for software maintenance, reuse, and quality assurance (Roy et al. 2009). However, code clone detection is challenging, as source code can be written in different languages, domains, and styles. In this paper, we argue that source code is inherently a graph, not a sequence, and that graph-based methods are more suitable for code clone detection than sequence-based methods. We compare the performance of two state-of-the-art models: CodeBERT (Feng et al. 2020), a sequence-based model, and CodeGraph (Yu et al. 2023), a graph-based model, on two benchmark data-sets: BCB (Svajlenko et al. 2014) and PoolC (PoolC no date). We show that CodeGraph outperforms CodeBERT on both data-sets, especially on cross-lingual code clones. To the best of our knowledge, this is the first work to demonstrate the superiority of graph-based methods over sequence-based methods on cross-lingual code clone detection.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16488</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16488</link>
            <enclosure url="https://arxiv.org/pdf/2312.16488"  type="application/pdf" />
            <author><![CDATA[Mohammed Ataaur Rahaman; Julia Ive]]></author>
        </item>
        <item>
            <title><![CDATA[Transfer and Alignment Network for Generalized Category Discovery]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16467">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16467">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16467">[Kimi]</a>
  <p>Generalized Category Discovery is a crucial real-world task. Despite the improved performance on known categories, current methods perform poorly on novel categories. We attribute the poor performance to two reasons: biased knowledge transfer between labeled and unlabeled data and noisy representation learning on the unlabeled data. To mitigate these two issues, we propose a Transfer and Alignment Network (TAN), which incorporates two knowledge transfer mechanisms to calibrate the biased knowledge and two feature alignment mechanisms to learn discriminative features. Specifically, we model different categories with prototypes and transfer the prototypes in labeled data to correct model bias towards known categories. On the one hand, we pull instances with known categories in unlabeled data closer to these prototypes to form more compact clusters and avoid boundary overlap between known and novel categories. On the other hand, we use these prototypes to calibrate noisy prototypes estimated from unlabeled data based on category similarities, which allows for more accurate estimation of prototypes for novel categories that can be used as reliable learning targets later. After knowledge transfer, we further propose two feature alignment mechanisms to acquire both instance- and category-level knowledge from unlabeled data by aligning instance features with both augmented features and the calibrated prototypes, which can boost model performance on both known and novel categories with less noise. Experiments on three benchmark datasets show that our model outperforms SOTA methods, especially on novel categories. Theoretical analysis is provided for an in-depth understanding of our model in general. Our code and data are available at https://github.com/Lackel/TAN.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16467</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16467</link>
            <enclosure url="https://arxiv.org/pdf/2312.16467"  type="application/pdf" />
            <author><![CDATA[Wenbin An; Feng Tian; Wenkai Shi; Yan Chen; Yaqiang Wu; Qianying Wang; Ping Chen]]></author>
        </item>
        <item>
            <title><![CDATA[Automating Knowledge Acquisition for Content-Centric Cognitive Agents Using LLMs]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16378">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16378">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16378">[Kimi]</a>
  <p>The paper describes a system that uses large language model (LLM) technology to support the automatic learning of new entries in an intelligent agent's semantic lexicon. The process is bootstrapped by an existing non-toy lexicon and a natural language generator that converts formal, ontologically-grounded representations of meaning into natural language sentences. The learning method involves a sequence of LLM requests and includes an automatic quality control step. To date, this learning method has been applied to learning multiword expressions whose meanings are equivalent to those of transitive verbs in the agent's lexicon. The experiment demonstrates the benefits of a hybrid learning architecture that integrates knowledge-based methods and resources with both traditional data analytics and LLMs.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16378</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16378</link>
            <enclosure url="https://arxiv.org/pdf/2312.16378"  type="application/pdf" />
            <author><![CDATA[Sanjay Oruganti; Sergei Nirenburg; Jesse English; Marjorie McShane]]></author>
        </item>
        <item>
            <title><![CDATA[LLM Polygraph: Uncovering LLMs' Factual Discernment through Intermediate Data Analysis]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16374">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16374">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16374">[Kimi]</a>
  <p>Large Language Models (LLMs) have revolutionized various domains with extensive knowledge and creative capabilities. However, a critical issue with LLMs is their tendency to produce outputs that diverge from factual reality. This phenomenon is particularly concerning in sensitive applications such as medical consultation and legal advice, where accuracy is paramount. In this paper, we introduce the LLM factoscope, a novel Siamese network-based model that leverages the inner states of LLMs for factual detection. Our investigation reveals distinguishable patterns in LLMs' inner states when generating factual versus non-factual content. We demonstrate the LLM factoscope's effectiveness across various architectures, achieving over 96% accuracy in factual detection. Our work opens a new avenue for utilizing LLMs' inner states for factual detection and encourages further exploration into LLMs' inner workings for enhanced reliability and transparency.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16374</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16374</link>
            <enclosure url="https://arxiv.org/pdf/2312.16374"  type="application/pdf" />
            <author><![CDATA[Jinwen He; Yujia Gong; Kai Chen; Zijin Lin; Chengan Wei; Yue Zhao]]></author>
        </item>
        <item>
            <title><![CDATA[Task Contamination: Language Models May Not Be Few-Shot Anymore]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16337">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16337">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16337">[Kimi]</a>
  <p>Large language models (LLMs) offer impressive performance in various zero-shot and few-shot tasks. However, their success in zero-shot and few-shot settings may be affected by task contamination, a potent

website/docs/routes/journal.mdx

Co-authored-by: Tony <TonyRL@users.noreply.github.com>

github-actions · 2023-12-29T14:23:34Z

Successfully generated as following:

http://localhost:1200/papers/arxiv/cs.CL - Success ✔️

<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"
>
    <channel>
        <title><![CDATA[Computation and Language | Cool Papers]]></title>
        <link>https://papers.cool/arxiv/cs.CL</link>
        <atom:link href="http://localhost:1200/papers/arxiv/cs.CL" rel="self" type="application/rss+xml" />
        <description><![CDATA[Computation and Language | Cool Papers - Immersive Paper Discovery - Made with love by RSSHub(https://github.com/DIYgod/RSSHub)]]></description>
        <generator>RSSHub</generator>
        <webMaster>i@diygod.me (DIYgod)</webMaster>
        <language>zh-cn</language>
        <lastBuildDate>Fri, 29 Dec 2023 14:23:33 GMT</lastBuildDate>
        <ttl>5</ttl>
        <item>
            <title><![CDATA[Do Androids Know They're Only Dreaming of Electric Sheep?]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17249">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17249">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17249">[Kimi]</a>
  <p>We design probes trained on the internal representations of a transformer language model that are predictive of its hallucinatory behavior on in-context generation tasks. To facilitate this detection, we create a span-annotated dataset of organic and synthetic hallucinations over several tasks. We find that probes trained on the force-decoded states of synthetic hallucinations are generally ecologically invalid in organic hallucination detection. Furthermore, hidden state information about hallucination appears to be task and distribution-dependent. Intrinsic and extrinsic hallucination saliency varies across layers, hidden state types, and tasks; notably, extrinsic hallucinations tend to be more salient in a transformer's internal representations. Outperforming multiple contemporary baselines, we show that probing is a feasible and efficient alternative to language model hallucination evaluation when model states are available.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17249</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17249</link>
            <enclosure url="https://arxiv.org/pdf/2312.17249"  type="application/pdf" />
            <author><![CDATA[Sky CH-Wang; Benjamin Van Durme; Jason Eisner; Chris Kedzie]]></author>
        </item>
        <item>
            <title><![CDATA[Learning to Generate Text in Arbitrary Writing Styles]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17242">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17242">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17242">[Kimi]</a>
  <p>Prior work in style-controlled text generation has focused on tasks such as emulating the style of prolific literary authors, producing formal or informal text, and the degree of toxicity of generated text. Plentiful demonstrations of these styles are available, and as a result modern language models are often able to emulate them, either via prompting or discriminative control. However, in applications such as writing assistants, it is desirable for language models to produce text in an author-specific style on the basis of a small writing sample. We find that instruction-tuned language models can struggle to reproduce author-specific style demonstrated in a prompt. Instead, we propose to guide a language model to generate text in a target style using contrastively-trained representations that capture stylometric features. A central challenge in doing so is that an author's writing is characterized by surprising token choices under a generic language model. To reconcile this tension, we combine generative re-scoring to achieve an author-specific model, with discriminative control to ensure style consistency at the sequence-level. The combination of these approaches is found to be particularly effective at adhering to an author-specific style in a variety of conditions, including unconditional generation and style transfer, and is applicable to any underlying language model without requiring fine-tuning.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17242</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17242</link>
            <enclosure url="https://arxiv.org/pdf/2312.17242"  type="application/pdf" />
            <author><![CDATA[Aleem Khan; Andrew Wang; Sophia Hager; Nicholas Andrews]]></author>
        </item>
        <item>
            <title><![CDATA[Virtual Scientific Companion for Synchrotron Beamlines: A Prototype]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17180">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17180">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17180">[Kimi]</a>
  <p>The extraordinarily high X-ray flux and specialized instrumentation at synchrotron beamlines have enabled versatile in-situ and high throughput studies that are impossible elsewhere. Dexterous and efficient control of experiments are thus crucial for efficient beamline operation. Artificial intelligence and machine learning methods are constantly being developed to enhance facility performance, but the full potential of these developments can only be reached with efficient human-computer-interaction. Natural language is the most intuitive and efficient way for humans to communicate. However, the low credibility and reproducibility of existing large language models and tools demand extensive development to be made for robust and reliable performance for scientific purposes. In this work, we introduce the prototype of virtual scientific companion (VISION) and demonstrate that it is possible to control basic beamline operations through natural language with open-source language model and the limited computational resources at beamline. The human-AI nature of VISION leverages existing automation systems and data framework at synchrotron beamlines.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17180</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17180</link>
            <enclosure url="https://arxiv.org/pdf/2312.17180"  type="application/pdf" />
            <author><![CDATA[Daniel Potemkin; Carlos Soto; Ruipeng Li; Kevin Yager; Esther Tsai]]></author>
        </item>
        <item>
            <title><![CDATA[Large Language Model for Causal Decision Making]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17122">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17122">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17122">[Kimi]</a>
  <p>Large Language Models (LLMs) have shown their success in language understanding and reasoning on general topics. However, their capability to inference based on user-specified structured data and knowledge in corpus-rare concepts like causal decision-making is still limited. In this work, we explore the possibility of fine-tuning an open-sourced LLM into LLM4Causal, which can identify the causal task, execute a corresponding function, and interpret its numerical results based on users' queries and the provided dataset. Meanwhile, we propose a data generation process for more controllable GPT prompting and present two instruction-tuning datasets: (1) Causal-Retrieval-Bench for causal problem identification and input parameter extraction for causal function calling and (2) Causal-Interpret-Bench for in-context causal interpretation. With three case studies, we showed that LLM4Causal can deliver end-to-end solutions for causal problems and provide easy-to-understand answers. Numerical studies also reveal that it has a remarkable ability to identify the correct causal task given a query.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17122</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17122</link>
            <enclosure url="https://arxiv.org/pdf/2312.17122"  type="application/pdf" />
            <author><![CDATA[Haitao Jiang; Lin Ge; Yuhe Gao; Jianian Wang; Rui Song]]></author>
        </item>
        <item>
            <title><![CDATA[Generative AI for Math: Part I -- MathPile: A Billion-Token-Scale Pretraining Corpus for Math]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17120">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17120">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17120">[Kimi]</a>
  <p>High-quality, large-scale corpora are the cornerstone of building foundation models. In this work, we introduce \textsc{MathPile}, a diverse and high-quality math-centric corpus comprising about 9.5 billion tokens. Throughout its creation, we adhered to the principle of ``\emph{less is more}'', firmly believing in the supremacy of data quality over quantity, even in the pre-training phase. Our meticulous data collection and processing efforts included a complex suite of preprocessing, prefiltering, language identification, cleaning, filtering, and deduplication, ensuring the high quality of our corpus. Furthermore, we performed data contamination detection on downstream benchmark test sets to eliminate duplicates. We hope our \textsc{MathPile} can help to enhance the mathematical reasoning abilities of language models. We plan to open-source different versions of \mathpile with the scripts used for processing, to facilitate future developments in this field.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17120</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17120</link>
            <enclosure url="https://arxiv.org/pdf/2312.17120"  type="application/pdf" />
            <author><![CDATA[Zengzhi Wang; Rui Xia; Pengfei Liu]]></author>
        </item>
        <item>
            <title><![CDATA[How Far Are We from Believable AI Agents? A Framework for Evaluating the Believability of Human Behavior Simulation]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17115">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17115">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17115">[Kimi]</a>
  <p>Human behavior simulation of AI agents necessitates the agents to possess a quality of believability, which is crucial as it facilitates users in establishing trust toward the agents and streamlines the fulfillment of the agents' goal. While recent advancements in Large Language Model (LLM) based agents have improved human behavior simulation, challenges inherent to LLMs (e.g., long context modeling) can undermine their believability. Consequently, evaluating AI agent believability becomes imperative. Unfortunately, prior research often neglects the negative impacts of LLM deficiencies. To address these gaps, we introduce two metrics for assessing LLM-based agent believability: consistency, and robustness, together with a benchmark, SimulateBench, with which, we evaluate the consistency and robustness of agents implemented with popular LLMs. We find that agents (i) struggle to accurately depict character information when presented with lengthy profile inputs; (ii) exhibit vulnerability to profile perturbations; and (iii) are significantly affected by certain key factors that impact their overall believability. Code and SimulateBench are public at https://github.com/GAIR-NLP/GPTMan.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17115</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17115</link>
            <enclosure url="https://arxiv.org/pdf/2312.17115"  type="application/pdf" />
            <author><![CDATA[Yang Xiao; Yi Cheng; Jinlan Fu; Jiashuo Wang; Wenjie Li; Pengfei Liu]]></author>
        </item>
        <item>
            <title><![CDATA[Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17080">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17080">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17080">[Kimi]</a>
  <p>In this work, we introduce a novel evaluation paradigm for Large Language Models, one that challenges them to engage in meta-reasoning. This approach addresses critical shortcomings in existing math problem-solving benchmarks, traditionally used to evaluate the cognitive capabilities of agents. Our paradigm shifts the focus from result-oriented assessments, which often overlook the reasoning process, to a more holistic evaluation that effectively differentiates the cognitive capabilities among models. For example, in our benchmark, GPT-4 demonstrates a performance ten times more accurate than GPT3-5. The significance of this new paradigm lies in its ability to reveal potential cognitive deficiencies in LLMs that current benchmarks, such as GSM8K, fail to uncover due to their saturation and lack of effective differentiation among varying reasoning abilities. Our comprehensive analysis includes several state-of-the-art math models from both open-source and closed-source communities, uncovering fundamental deficiencies in their training and evaluation approaches. This paper not only advocates for a paradigm shift in the assessment of LLMs but also contributes to the ongoing discourse on the trajectory towards Artificial General Intelligence (AGI). By promoting the adoption of meta-reasoning evaluation methods similar to ours, we aim to facilitate a more accurate assessment of the true cognitive abilities of LLMs.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17080</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17080</link>
            <enclosure url="https://arxiv.org/pdf/2312.17080"  type="application/pdf" />
            <author><![CDATA[Zhongshen Zeng; Pengguang Chen; Haiyun Jiang; Jiaya Jia]]></author>
        </item>
        <item>
            <title><![CDATA[Improving In-context Learning via Bidirectional Alignment]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17055">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17055">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17055">[Kimi]</a>
  <p>Large language models (LLMs) have shown impressive few-shot generalization on many tasks via in-context learning (ICL). Despite their success in showing such emergent abilities, the scale and complexity of larger models also lead to unprecedentedly high computational demands and deployment challenges. In reaction, researchers explore transferring the powerful capabilities of larger models to more efficient and compact models by typically aligning the output of smaller models with that of larger models. Existing methods either train smaller models on the generated outputs of larger models or to imitate their token-level probability distributions. However, these distillation methods pay little to no attention to the input part, which also plays a crucial role in ICL. Based on the finding that the performance of ICL is highly sensitive to the selection of demonstration examples, we propose Bidirectional Alignment (BiAlign) to fully leverage the models' preferences for ICL examples to improve the ICL abilities of smaller models. Specifically, we introduce the alignment of input preferences between smaller and larger models by incorporating a novel ranking loss, in addition to aligning the token-level output distribution. With extensive experiments and analysis, we demonstrate that BiAlign can consistently outperform existing baselines on a variety of tasks including language understanding, reasoning, and coding.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17055</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17055</link>
            <enclosure url="https://arxiv.org/pdf/2312.17055"  type="application/pdf" />
            <author><![CDATA[Chengwei Qin; Wenhan Xia; Fangkai Jiao; Shafiq Joty]]></author>
        </item>
        <item>
            <title><![CDATA[Length Extrapolation of Transformers: A Survey from the Perspective of Position Encoding]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17044">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17044">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17044">[Kimi]</a>
  <p>Transformer has taken the natural language processing (NLP) field by storm since birth, owing to its superior ability to model complex dependencies in sequences. Despite the great success of pretrained language models (PLMs) based on Transformer across almost all NLP tasks, they all suffer from a preset length limit and thus can hardly extend this success to longer sequences beyond seen data, namely the length extrapolation problem. Length extrapolation has aroused great interest among researchers, as it is the core feature of human language capacity. To enhance length extrapolation of Transformers, a plethora of methods have been proposed, mostly focusing on extrapolatable position encodings. In this article, we provide an organized and systematical review of these research efforts in a unified notation from a position encoding perspective, aiming to enable the reader to gain a deep understanding of existing methods and provide stimuli for future research.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17044</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17044</link>
            <enclosure url="https://arxiv.org/pdf/2312.17044"  type="application/pdf" />
            <author><![CDATA[Liang Zhao; Xiaocheng Feng; Xiachong Feng; Bin Qin; Ting Liu]]></author>
        </item>
        <item>
            <title><![CDATA[Experiential Co-Learning of Software-Developing Agents]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.17025">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.17025">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.17025">[Kimi]</a>
  <p>Recent advancements in large language models (LLMs) have brought significant changes to various dimains, especially through LLM-driven autonomous agents. These agents are now capable of collaborating seamlessly, splitting tasks and enhancing accuracy, thus minimizing the need for human involvement. However, these agents often approach a diverse range of tasks in isolation, without benefiting from past experiences. This isolation can lead to repeated mistakes and inefficient trials in task solving. To this end, this paper introduces Experiential Co-Learning, a novel framework in which instructor and assistant agents gather shortcut-oriented experiences from their historical trajectories and use these past experiences for mutual reasoning. This paradigm, enriched with previous experiences, equips agents to more effectively address unseen tasks.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.17025</guid>
            <link>https://papers.cool/arxiv/kimi/2312.17025</link>
            <enclosure url="https://arxiv.org/pdf/2312.17025"  type="application/pdf" />
            <author><![CDATA[Chen Qian; Yufan Dang; Jiahao Li; Wei Liu; Weize Chen; Cheng Yang; Zhiyuan Liu; Maosong Sun]]></author>
        </item>
        <item>
            <title><![CDATA[Few-shot learning for automated content analysis: Efficient coding of arguments and claims in the debate on arms deliveries to Ukraine]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16975">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16975">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16975">[Kimi]</a>
  <p>Pre-trained language models (PLM) based on transformer neural networks developed in the field of natural language processing (NLP) offer great opportunities to improve automatic content analysis in communication science, especially for the coding of complex semantic categories in large datasets via supervised machine learning. However, three characteristics so far impeded the widespread adoption of the methods in the applying disciplines: the dominance of English language models in NLP research, the necessary computing resources, and the effort required to produce training data to fine-tune PLMs. In this study, we address these challenges by using a multilingual transformer model in combination with the adapter extension to transformers, and few-shot learning methods. We test our approach on a realistic use case from communication science to automatically detect claims and arguments together with their stance in the German news debate on arms deliveries to Ukraine. In three experiments, we evaluate (1) data preprocessing strategies and model variants for this task, (2) the performance of different few-shot learning methods, and (3) how well the best setup performs on varying training set sizes in terms of validity, reliability, replicability and reproducibility of the results. We find that our proposed combination of transformer adapters with pattern exploiting training provides a parameter-efficient and easily shareable alternative to fully fine-tuning PLMs. It performs on par in terms of validity, while overall, provides better properties for application in communication studies. The results also show that pre-fine-tuning for a task on a near-domain dataset leads to substantial improvement, in particular in the few-shot setting. Further, the results indicate that it is useful to bias the dataset away from the viewpoints of specific prominent individuals.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16975</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16975</link>
            <enclosure url="https://arxiv.org/pdf/2312.16975"  type="application/pdf" />
            <author><![CDATA[Jonas Rieger; Kostiantyn Yanchenko; Mattes Ruckdeschel; Gerret von Nordheim; Katharina Kleinen-von Königslöw; Gregor Wiedemann]]></author>
        </item>
        <item>
            <title><![CDATA[Unified Lattice Graph Fusion for Chinese Named Entity Recognition]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16917">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16917">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16917">[Kimi]</a>
  <p>Integrating lexicon into character-level sequence has been proven effective to leverage word boundary and semantic information in Chinese named entity recognition (NER). However, prior approaches usually utilize feature weighting and position coupling to integrate word information, but ignore the semantic and contextual correspondence between the fine-grained semantic units in the character-word space. To solve this issue, we propose a Unified Lattice Graph Fusion (ULGF) approach for Chinese NER. ULGF can explicitly capture various semantic and boundary relations across different semantic units with the adjacency matrix by converting the lattice structure into a unified graph. We stack multiple graph-based intra-source self-attention and inter-source cross-gating fusion layers that iteratively carry out semantic interactions to learn node representations. To alleviate the over-reliance on word information, we further propose to leverage lexicon entity classification as an auxiliary task. Experiments on four Chinese NER benchmark datasets demonstrate the superiority of our ULGF approach.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16917</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16917</link>
            <enclosure url="https://arxiv.org/pdf/2312.16917"  type="application/pdf" />
            <author><![CDATA[Dixiang Zhang; Junyu Lu; Pingjian Zhang]]></author>
        </item>
        <item>
            <title><![CDATA[Spike No More: Stabilizing the Pre-training of Large Language Models]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16903">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16903">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16903">[Kimi]</a>
  <p>The loss spike often occurs during pre-training of a large language model. The spikes degrade the performance of a large language model, and sometimes ruin the pre-training. Since the pre-training needs a vast computational budget, we should avoid such spikes. To investigate a cause of loss spikes, we focus on gradients of internal layers in this study. Through theoretical analyses, we introduce two causes of the exploding gradients, and provide requirements to prevent the explosion. In addition, we introduce the combination of the initialization method and a simple modification to embeddings as a method to satisfy the requirements. We conduct various experiments to verify our theoretical analyses empirically. Experimental results indicate that the combination is effective in preventing spikes during pre-training.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16903</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16903</link>
            <enclosure url="https://arxiv.org/pdf/2312.16903"  type="application/pdf" />
            <author><![CDATA[Sho Takase; Shun Kiyono; Sosuke Kobayashi; Jun Suzuki]]></author>
        </item>
        <item>
            <title><![CDATA[BBScore: A Brownian Bridge Based Metric for Assessing Text Coherence]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16893">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16893">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16893">[Kimi]</a>
  <p>Measuring the coherence of text is a vital aspect of evaluating the quality of written content. Recent advancements in neural coherence modeling have demonstrated their efficacy in capturing entity coreference and discourse relations, thereby enhancing coherence evaluation. However, many existing methods heavily depend on static embeddings or focus narrowly on nearby context, constraining their capacity to measure the overarching coherence of long texts. In this paper, we posit that coherent texts inherently manifest a sequential and cohesive interplay among sentences, effectively conveying the central theme, purpose, or standpoint. To explore this abstract relationship, we introduce the "BBScore," a novel reference-free metric grounded in Brownian bridge theory for assessing text coherence. Our findings showcase that when synergized with a simple additional classification component, this metric attains a performance level comparable to state-of-the-art techniques on standard artificial discrimination tasks. We also establish in downstream tasks that this metric effectively differentiates between human-written documents and text generated by large language models under a specific domain. Furthermore, we illustrate the efficacy of this approach in detecting written styles attributed to diverse large language models, underscoring its potential for generalizability. In summary, we present a novel Brownian bridge coherence metric capable of measuring both local and global text coherence, while circumventing the need for end-to-end model training. This flexibility allows for its application in various downstream tasks.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16893</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16893</link>
            <enclosure url="https://arxiv.org/pdf/2312.16893"  type="application/pdf" />
            <author><![CDATA[Zhecheng Sheng; Tianhao Zhang; Chen Jiang; Dongyeop Kang]]></author>
        </item>
        <item>
            <title><![CDATA[OmniDialog: An Omnipotent Pre-training Model for Task-Oriented Dialogue System]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16864">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16864">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16864">[Kimi]</a>
  <p>Pre-trained conversation models (PCMs) have demonstrated remarkable results in task-oriented dialogue (TOD) systems. Many PCMs focus predominantly on dialogue management tasks like dialogue state tracking, dialogue generation tasks like response generation, or both. However, the existing PCMs seldom consider dialogue comprehension tasks, such as dialogue question answering and summarization tasks. These tasks allow PCMs to glean dialogue context from various angles. This observation naturally raises the question: Can the performance of downstream dialogue tasks be enhanced if a PCM is pre-trained on dialogue management, generation, and comprehension tasks? To investigate this, we proposed an Omnipotent Dialogue pre-training model (OmniDialog). It unifies these three dialogue tasks into a monolithic framework by multi-task learning, fostering inter-task communication. The pre-training corpus of OmniDialog spans $\mathbf{7}$ dialogue-focused tasks, drawing from $\mathbf{15}$ datasets and encompassing over $\mathbf{3.2}$ million dialogue utterances. To our knowledge, OmniDialog is a pioneering PCM pre-trained across dialogue management, generation, and comprehension domains. We evaluated its performance across four tasks: dialogue summarization, end-to-end dialogue modeling, dialogue state tracking, and intent classification. The results underscore its efficacy in domain transfer learning, low-resource, and full-dataset scenarios. Furthermore, to glean a nuanced understanding of OmniDialog's strengths and potential pitfalls, we designed a fine-grained analysis framework for dialogue-centric tasks. Experimental results show that the OmniDialog is good at hard samples, such as long dialogues and lengthy responses.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16864</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16864</link>
            <enclosure url="https://arxiv.org/pdf/2312.16864"  type="application/pdf" />
            <author><![CDATA[Mingtao Yang; See-Kiong Ng; Jinlan Fu]]></author>
        </item>
        <item>
            <title><![CDATA[Evaluating the Performance of Large Language Models for Spanish Language in Undergraduate Admissions Exams]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16845">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16845">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16845">[Kimi]</a>
  <p>This study evaluates the performance of large language models, specifically GPT-3.5 and BARD (supported by Gemini Pro model), in undergraduate admissions exams proposed by the National Polytechnic Institute in Mexico. The exams cover Engineering/Mathematical and Physical Sciences, Biological and Medical Sciences, and Social and Administrative Sciences. Both models demonstrated proficiency, exceeding the minimum acceptance scores for respective academic programs to up to 75% for some academic programs. GPT-3.5 outperformed BARD in Mathematics and Physics, while BARD performed better in History and questions related to factual information. Overall, GPT-3.5 marginally surpassed BARD with scores of 60.94% and 60.42%, respectively.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16845</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16845</link>
            <enclosure url="https://arxiv.org/pdf/2312.16845"  type="application/pdf" />
            <author><![CDATA[Sabino Miranda; Obdulia Pichardo-Lagunas; Bella Martínez-Seis; Pierre Baldi]]></author>
        </item>
        <item>
            <title><![CDATA[Hiding in Plain Sight: Towards the Science of Linguistic Steganography]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16840">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16840">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16840">[Kimi]</a>
  <p>Covert communication (also known as steganography) is the practice of concealing a secret inside an innocuous-looking public object (cover) so that the modified public object (covert code) makes sense to everyone but only someone who knows the code can extract the secret (message). Linguistic steganography is the practice of encoding a secret message in natural language text such as spoken conversation or short public communications such as tweets.. While ad hoc methods for covert communications in specific domains exist ( JPEG images, Chinese poetry, etc), there is no general model for linguistic steganography specifically. We present a novel mathematical formalism for creating linguistic steganographic codes, with three parameters: Decodability (probability that the receiver of the coded message will decode the cover correctly), density (frequency of code words in a cover code), and detectability (probability that an attacker can tell the difference between an untampered cover compared to its steganized version). Verbal or linguistic steganography is most challenging because of its lack of artifacts to hide the secret message in. We detail a practical construction in Python of a steganographic code for Tweets using inserted words to encode hidden digits while using n-gram frequency distortion as the measure of detectability of the insertions. Using the publicly accessible Stanford Sentiment Analysis dataset we implemented the tweet steganization scheme -- a codeword (an existing word in the data set) inserted in random positions in random existing tweets to find the tweet that has the least possible n-gram distortion. We argue that this approximates KL distance in a localized manner at low cost and thus we get a linguistic steganography scheme that is both formal and practical and permits a tradeoff between codeword density and detectability of the covert message.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16840</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16840</link>
            <enclosure url="https://arxiv.org/pdf/2312.16840"  type="application/pdf" />
            <author><![CDATA[Leela Raj-Sankar; S. Raj Rajagopalan]]></author>
        </item>
        <item>
            <title><![CDATA[Adversarial Representation with Intra-Modal and Inter-Modal Graph Contrastive Learning for Multimodal Emotion Recognition]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16778">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16778">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16778">[Kimi]</a>
  <p>With the release of increasing open-source emotion recognition datasets on social media platforms and the rapid development of computing resources, multimodal emotion recognition tasks (MER) have begun to receive widespread research attention. The MER task extracts and fuses complementary semantic information from different modalities, which can classify the speaker's emotions. However, the existing feature fusion methods have usually mapped the features of different modalities into the same feature space for information fusion, which can not eliminate the heterogeneity between different modalities. Therefore, it is challenging to make the subsequent emotion class boundary learning. To tackle the above problems, we have proposed a novel Adversarial Representation with Intra-Modal and Inter-Modal Graph Contrastive for Multimodal Emotion Recognition (AR-IIGCN) method. Firstly, we input video, audio, and text features into a multi-layer perceptron (MLP) to map them into separate feature spaces. Secondly, we build a generator and a discriminator for the three modal features through adversarial representation, which can achieve information interaction between modalities and eliminate heterogeneity among modalities. Thirdly, we introduce contrastive graph representation learning to capture intra-modal and inter-modal complementary semantic information and learn intra-class and inter-class boundary information of emotion categories. Specifically, we construct a graph structure for three modal features and perform contrastive representation learning on nodes with different emotions in the same modality and the same emotion in different modalities, which can improve the feature representation ability of nodes. Extensive experimental works show that the ARL-IIGCN method can significantly improve emotion recognition accuracy on IEMOCAP and MELD datasets.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16778</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16778</link>
            <enclosure url="https://arxiv.org/pdf/2312.16778"  type="application/pdf" />
            <author><![CDATA[Yuntao Shou; Tao Meng; Wei Ai; Keqin Li]]></author>
        </item>
        <item>
            <title><![CDATA[Graph Neural Networks for Antisocial Behavior Detection on Twitter]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16755">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16755">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16755">[Kimi]</a>
  <p>Social media resurgence of antisocial behavior has exerted a downward spiral on stereotypical beliefs, and hateful comments towards individuals and social groups, as well as false or distorted news. The advances in graph neural networks employed on massive quantities of graph-structured data raise high hopes for the future of mediating communication on social media platforms. An approach based on graph convolutional data was employed to better capture the dependencies between the heterogeneous types of data. Utilizing past and present experiences on the topic, we proposed and evaluated a graph-based approach for antisocial behavior detection, with general applicability that is both language- and context-independent. In this research, we carried out an experimental validation of our graph-based approach on several PAN datasets provided as part of their shared tasks, that enable the discussion of the results obtained by the proposed solution.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16755</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16755</link>
            <enclosure url="https://arxiv.org/pdf/2312.16755"  type="application/pdf" />
            <author><![CDATA[Martina Toshevska; Slobodan Kalajdziski; Sonja Gievska]]></author>
        </item>
        <item>
            <title><![CDATA[A Reversible Perspective on Petri Nets and Event Structures]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16714">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16714">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16714">[Kimi]</a>
  <p>Event structures have emerged as a foundational model for concurrent computation, explaining computational processes by outlining the events and the relationships that dictate their execution. They play a pivotal role in the study of key aspects of concurrent computation models, such as causality and independence, and have found applications across a broad range of languages and models, spanning realms like persistence, probabilities, and quantum computing. Recently, event structures have been extended to address reversibility, where computational processes can undo previous computations. In this context, reversible event structures provide abstract representations of processes capable of both forward and backward steps in a computation. Since their introduction, event structures have played a crucial role in bridging operational models, traditionally exemplified by Petri nets and process calculi, with denotational ones, i.e., algebraic domains. In this context, we revisit the standard connection between Petri nets and event structures under the lenses of reversibility. Specifically, we introduce a subset of contextual Petri nets, dubbed reversible causal nets, that precisely correspond to reversible prime event structures. The distinctive feature of reversible causal nets lies in deriving causality from inhibitor arcs, departing from the conventional dependence on the overlap between the post and preset of transitions. In this way, we are able to operationally explain the full model of reversible prime event structures.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16714</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16714</link>
            <enclosure url="https://arxiv.org/pdf/2312.16714"  type="application/pdf" />
            <author><![CDATA[Hernán Melgratti; Claudio Antares Mezzina; G. Michele Pinna]]></author>
        </item>
        <item>
            <title><![CDATA[Rethinking Tabular Data Understanding with Large Language Models]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16702">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16702">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16702">[Kimi]</a>
  <p>Large Language Models (LLMs) have shown to be capable of various tasks, yet their capability in interpreting and reasoning over tabular data remains an underexplored area. In this context, this study investigates from three core perspectives: the robustness of LLMs to structural perturbations in tables, the comparative analysis of textual and symbolic reasoning on tables, and the potential of boosting model performance through the aggregation of multiple reasoning pathways. We discover that structural variance of tables presenting the same content reveals a notable performance decline, particularly in symbolic reasoning tasks. This prompts the proposal of a method for table structure normalization. Moreover, textual reasoning slightly edges out symbolic reasoning, and a detailed error analysis reveals that each exhibits different strengths depending on the specific tasks. Notably, the aggregation of textual and symbolic reasoning pathways, bolstered by a mix self-consistency mechanism, resulted in achieving SOTA performance, with an accuracy of 73.6% on WIKITABLEQUESTIONS, representing a substantial advancement over previous existing table processing paradigms of LLMs.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16702</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16702</link>
            <enclosure url="https://arxiv.org/pdf/2312.16702"  type="application/pdf" />
            <author><![CDATA[Tianyang Liu; Fei Wang; Muhao Chen]]></author>
        </item>
        <item>
            <title><![CDATA[Some things are more CRINGE than others: Preference Optimization with the Pairwise Cringe Loss]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16682">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16682">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16682">[Kimi]</a>
  <p>Practitioners commonly align large language models using pairwise preferences, i.e., given labels of the type response A is preferred to response B for a given input. Perhaps less commonly, methods have also been developed for binary feedback, i.e. training models given labels of type response A is good or bad. We show how an existing performant binary feedback method, the Cringe Loss (Adolphs et al., 2022), can be generalized to the pairwise preference setting using a simple soft margin extension. Pairwise Cringe Loss is straightforward to implement and efficient to train, and we find it outperforms state-of-the-art preference optimization algorithms such as PPO and DPO on the AlpacaFarm benchmark.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16682</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16682</link>
            <enclosure url="https://arxiv.org/pdf/2312.16682"  type="application/pdf" />
            <author><![CDATA[Jing Xu; Andrew Lee; Sainbayar Sukhbaatar; Jason Weston]]></author>
        </item>
        <item>
            <title><![CDATA[A Large Language Model-based Computational Approach to Improve Identity-Related Write-Ups]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16659">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16659">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16659">[Kimi]</a>
  <p>Creating written products is essential to modern life, including writings about one's identity and personal experiences. However, writing is often a difficult activity that requires extensive effort to frame the central ideas, the pursued approach to communicate the central ideas, e.g., using analogies, metaphors, or other possible means, the needed presentation structure, and the actual verbal expression. Large Language Models, a recently emerged approach in Machine Learning, can offer a significant help in reducing the effort and improving the quality of written products. This paper proposes a new computational approach to explore prompts that given as inputs to a Large Language Models can generate cues to improve the considered written products. Two case studies on improving write-ups, one based on an analogy and one on a metaphor, are also presented in the paper.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16659</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16659</link>
            <enclosure url="https://arxiv.org/pdf/2312.16659"  type="application/pdf" />
            <author><![CDATA[Alex Doboli]]></author>
        </item>
        <item>
            <title><![CDATA[Make BERT-based Chinese Spelling Check Model Enhanced by Layerwise Attention and Gaussian Mixture Model]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16623">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16623">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16623">[Kimi]</a>
  <p>BERT-based models have shown a remarkable ability in the Chinese Spelling Check (CSC) task recently. However, traditional BERT-based methods still suffer from two limitations. First, although previous works have identified that explicit prior knowledge like Part-Of-Speech (POS) tagging can benefit in the CSC task, they neglected the fact that spelling errors inherent in CSC data can lead to incorrect tags and therefore mislead models. Additionally, they ignored the correlation between the implicit hierarchical information encoded by BERT's intermediate layers and different linguistic phenomena. This results in sub-optimal accuracy. To alleviate the above two issues, we design a heterogeneous knowledge-infused framework to strengthen BERT-based CSC models. To incorporate explicit POS knowledge, we utilize an auxiliary task strategy driven by Gaussian mixture model. Meanwhile, to incorporate implicit hierarchical linguistic knowledge within the encoder, we propose a novel form of n-gram-based layerwise self-attention to generate a multilayer representation. Experimental results show that our proposed framework yields a stable performance boost over four strong baseline models and outperforms the previous state-of-the-art methods on two datasets.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16623</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16623</link>
            <enclosure url="https://arxiv.org/pdf/2312.16623"  type="application/pdf" />
            <author><![CDATA[Yongchang Cao; Liang He; Zhen Wu; Xinyu Dai]]></author>
        </item>
        <item>
            <title><![CDATA[Relationship between auditory and semantic entrainment using Deep Neural Networks (DNN)]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16599">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16599">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16599">[Kimi]</a>
  <p>The tendency of people to engage in similar, matching, or synchronized behaviour when interacting is known as entrainment. Many studies examined linguistic (syntactic and lexical structures) and paralinguistic (pitch, intensity) entrainment, but less attention was given to finding the relationship between them. In this study, we utilized state-of-the-art DNN embeddings such as BERT and TRIpLet Loss network (TRILL) vectors to extract features for measuring semantic and auditory similarities of turns within dialogues in two comparable spoken corpora of two different languages. We found people's tendency to entrain on semantic features more when compared to auditory features. Additionally, we found that entrainment in semantic and auditory linguistic features are positively correlated. The findings of this study might assist in implementing the mechanism of entrainment in human-machine interaction (HMI).</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16599</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16599</link>
            <enclosure url="https://arxiv.org/pdf/2312.16599"  type="application/pdf" />
            <author><![CDATA[Jay Kejriwal; Štefan Beňuš]]></author>
        </item>
        <item>
            <title><![CDATA[A proposed new metric for the conceptual diversity of a text]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16548">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16548">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16548">[Kimi]</a>
  <p>A word may contain one or more hidden concepts. While the "animal" word evokes many images in our minds and encapsulates many concepts (birds, dogs, cats, crocodiles, etc.), the `parrot' word evokes a single image (a colored bird with a short, hooked beak and the ability to mimic sounds). In spoken or written texts, we use some words in a general sense and some in a detailed way to point to a specific object. Until now, a text's conceptual diversity value cannot be determined using a standard and precise technique. This research contributes to the natural language processing field of AI by offering a standardized method and a generic metric for evaluating and comparing concept diversity in different texts and domains. It also contributes to the field of semantic research of languages. If we give examples for the diversity score of two sentences, "He discovered an unknown entity." has a high conceptual diversity score (16.6801), and "The endoplasmic reticulum forms a series of flattened sacs within the cytoplasm of eukaryotic cells." sentence has a low conceptual diversity score which is 3.9068.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16548</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16548</link>
            <enclosure url="https://arxiv.org/pdf/2312.16548"  type="application/pdf" />
            <author><![CDATA[İlknur Dönmez Phd; Mehmet Haklıdır Phd]]></author>
        </item>
        <item>
            <title><![CDATA[S2M: Converting Single-Turn to Multi-Turn Datasets for Conversational Question Answering]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16511">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16511">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16511">[Kimi]</a>
  <p>Supplying data augmentation to conversational question answering (CQA) can effectively improve model performance. However, there is less improvement from single-turn datasets in CQA due to the distribution gap between single-turn and multi-turn datasets. On the other hand, while numerous single-turn datasets are available, we have not utilized them effectively. To solve this problem, we propose a novel method to convert single-turn datasets to multi-turn datasets. The proposed method consists of three parts, namely, a QA pair Generator, a QA pair Reassembler, and a question Rewriter. Given a sample consisting of context and single-turn QA pairs, the Generator obtains candidate QA pairs and a knowledge graph based on the context. The Reassembler utilizes the knowledge graph to get sequential QA pairs, and the Rewriter rewrites questions from a conversational perspective to obtain a multi-turn dataset S2M. Our experiments show that our method can synthesize effective training resources for CQA. Notably, S2M ranks 1st place on the QuAC leaderboard at the time of submission (Aug 24th, 2022).</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16511</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16511</link>
            <enclosure url="https://arxiv.org/pdf/2312.16511"  type="application/pdf" />
            <author><![CDATA[Baokui Li; Sen Zhang; Wangshu Zhang; Yicheng Chen; Changlin Yang; Sen Hu; Teng Xu; Siye liu; Jiwei Li]]></author>
        </item>
        <item>
            <title><![CDATA[Understanding News Creation Intents: Frame, Dataset, and Method]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16490">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16490">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16490">[Kimi]</a>
  <p>As the disruptive changes in the media economy and the proliferation of alternative news media outlets, news intent has progressively deviated from ethical standards that serve the public interest. News intent refers to the purpose or intention behind the creation of a news article. While the significance of research on news intent has been widely acknowledged, the absence of a systematic news intent understanding framework hinders further exploration of news intent and its downstream applications. To bridge this gap, we propose News INTent (NINT) frame, the first component-aware formalism for understanding the news creation intent based on research in philosophy, psychology, and cognitive science. Within this frame, we define the news intent identification task and provide a benchmark dataset with fine-grained labels along with an efficient benchmark method. Experiments demonstrate that NINT is beneficial in both the intent identification task and downstream tasks that demand a profound understanding of news. This work marks a foundational step towards a more systematic exploration of news creation intents.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16490</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16490</link>
            <enclosure url="https://arxiv.org/pdf/2312.16490"  type="application/pdf" />
            <author><![CDATA[Zhengjia Wang; Danding Wang; Qiang Sheng; Juan Cao; Silong Su; Yifan Sun; Beizhe Hu; Siyuan Ma]]></author>
        </item>
        <item>
            <title><![CDATA[Source Code is a Graph, Not a Sequence: A Cross-Lingual Perspective on Code Clone Detection]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16488">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16488">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16488">[Kimi]</a>
  <p>Source code clone detection is the task of finding code fragments that have the same or similar functionality, but may differ in syntax or structure. This task is important for software maintenance, reuse, and quality assurance (Roy et al. 2009). However, code clone detection is challenging, as source code can be written in different languages, domains, and styles. In this paper, we argue that source code is inherently a graph, not a sequence, and that graph-based methods are more suitable for code clone detection than sequence-based methods. We compare the performance of two state-of-the-art models: CodeBERT (Feng et al. 2020), a sequence-based model, and CodeGraph (Yu et al. 2023), a graph-based model, on two benchmark data-sets: BCB (Svajlenko et al. 2014) and PoolC (PoolC no date). We show that CodeGraph outperforms CodeBERT on both data-sets, especially on cross-lingual code clones. To the best of our knowledge, this is the first work to demonstrate the superiority of graph-based methods over sequence-based methods on cross-lingual code clone detection.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16488</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16488</link>
            <enclosure url="https://arxiv.org/pdf/2312.16488"  type="application/pdf" />
            <author><![CDATA[Mohammed Ataaur Rahaman; Julia Ive]]></author>
        </item>
        <item>
            <title><![CDATA[Transfer and Alignment Network for Generalized Category Discovery]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16467">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16467">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16467">[Kimi]</a>
  <p>Generalized Category Discovery is a crucial real-world task. Despite the improved performance on known categories, current methods perform poorly on novel categories. We attribute the poor performance to two reasons: biased knowledge transfer between labeled and unlabeled data and noisy representation learning on the unlabeled data. To mitigate these two issues, we propose a Transfer and Alignment Network (TAN), which incorporates two knowledge transfer mechanisms to calibrate the biased knowledge and two feature alignment mechanisms to learn discriminative features. Specifically, we model different categories with prototypes and transfer the prototypes in labeled data to correct model bias towards known categories. On the one hand, we pull instances with known categories in unlabeled data closer to these prototypes to form more compact clusters and avoid boundary overlap between known and novel categories. On the other hand, we use these prototypes to calibrate noisy prototypes estimated from unlabeled data based on category similarities, which allows for more accurate estimation of prototypes for novel categories that can be used as reliable learning targets later. After knowledge transfer, we further propose two feature alignment mechanisms to acquire both instance- and category-level knowledge from unlabeled data by aligning instance features with both augmented features and the calibrated prototypes, which can boost model performance on both known and novel categories with less noise. Experiments on three benchmark datasets show that our model outperforms SOTA methods, especially on novel categories. Theoretical analysis is provided for an in-depth understanding of our model in general. Our code and data are available at https://github.com/Lackel/TAN.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16467</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16467</link>
            <enclosure url="https://arxiv.org/pdf/2312.16467"  type="application/pdf" />
            <author><![CDATA[Wenbin An; Feng Tian; Wenkai Shi; Yan Chen; Yaqiang Wu; Qianying Wang; Ping Chen]]></author>
        </item>
        <item>
            <title><![CDATA[Automating Knowledge Acquisition for Content-Centric Cognitive Agents Using LLMs]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16378">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16378">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16378">[Kimi]</a>
  <p>The paper describes a system that uses large language model (LLM) technology to support the automatic learning of new entries in an intelligent agent's semantic lexicon. The process is bootstrapped by an existing non-toy lexicon and a natural language generator that converts formal, ontologically-grounded representations of meaning into natural language sentences. The learning method involves a sequence of LLM requests and includes an automatic quality control step. To date, this learning method has been applied to learning multiword expressions whose meanings are equivalent to those of transitive verbs in the agent's lexicon. The experiment demonstrates the benefits of a hybrid learning architecture that integrates knowledge-based methods and resources with both traditional data analytics and LLMs.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16378</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16378</link>
            <enclosure url="https://arxiv.org/pdf/2312.16378"  type="application/pdf" />
            <author><![CDATA[Sanjay Oruganti; Sergei Nirenburg; Jesse English; Marjorie McShane]]></author>
        </item>
        <item>
            <title><![CDATA[LLM Polygraph: Uncovering LLMs' Factual Discernment through Intermediate Data Analysis]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16374">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16374">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16374">[Kimi]</a>
  <p>Large Language Models (LLMs) have revolutionized various domains with extensive knowledge and creative capabilities. However, a critical issue with LLMs is their tendency to produce outputs that diverge from factual reality. This phenomenon is particularly concerning in sensitive applications such as medical consultation and legal advice, where accuracy is paramount. In this paper, we introduce the LLM factoscope, a novel Siamese network-based model that leverages the inner states of LLMs for factual detection. Our investigation reveals distinguishable patterns in LLMs' inner states when generating factual versus non-factual content. We demonstrate the LLM factoscope's effectiveness across various architectures, achieving over 96% accuracy in factual detection. Our work opens a new avenue for utilizing LLMs' inner states for factual detection and encourages further exploration into LLMs' inner workings for enhanced reliability and transparency.</p>
]]></description>
            <pubDate>Thu, 28 Dec 2023 16:00:00 GMT</pubDate>
            <guid isPermaLink="false">https://papers.cool/arxiv/cs.CL#2312.16374</guid>
            <link>https://papers.cool/arxiv/kimi/2312.16374</link>
            <enclosure url="https://arxiv.org/pdf/2312.16374"  type="application/pdf" />
            <author><![CDATA[Jinwen He; Yujia Gong; Kai Chen; Zijin Lin; Chengan Wei; Yue Zhao]]></author>
        </item>
        <item>
            <title><![CDATA[Task Contamination: Language Models May Not Be Few-Shot Anymore]]></title>
            <description><![CDATA[<a href="https://arxiv.org/pdf/2312.16337">[PDF]</a>
  <a href="https://arxiv.org/abs/2312.16337">[Site]</a>
  <a href="https://papers.cool/arxiv/kimi/2312.16337">[Kimi]</a>
  <p>Large language models (LLMs) offer impressive performance in various zero-shot and few-shot tasks. However, their success in zero-shot and few-shot settings may be affected by task contamination, a potent

* feat(route): add 米课圈精华 (#14010) * feat(route): add 米课圈精华 * fix typo * fix: radar.js with type errors * fix: radar.js with type errors * fix(route): Readhub (#14013) * fix(route): Readhub * fix typo * chore(deps-dev): bump @types/react from 18.2.42 to 18.2.43 in /website (#14018) Bumps [@types/react](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/react) from 18.2.42 to 18.2.43. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/react) --- updated-dependencies: - dependency-name: "@types/react" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump prettier from 3.1.0 to 3.1.1 (#14016) * chore(deps-dev): bump prettier from 3.1.0 to 3.1.1 Bumps [prettier](https://github.com/prettier/prettier) from 3.1.0 to 3.1.1. - [Release notes](https://github.com/prettier/prettier/releases) - [Changelog](https://github.com/prettier/prettier/blob/main/CHANGELOG.md) - [Commits](https://github.com/prettier/prettier/compare/3.1.0...3.1.1) --- updated-dependencies: - dependency-name: prettier dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @stylistic/eslint-plugin-js from 1.5.0 to 1.5.1 (#14015) * chore(deps-dev): bump @stylistic/eslint-plugin-js from 1.5.0 to 1.5.1 Bumps [@stylistic/eslint-plugin-js](https://github.com/eslint-stylistic/eslint-stylistic/tree/HEAD/packages/eslint-plugin-js) from 1.5.0 to 1.5.1. - [Release notes](https://github.com/eslint-stylistic/eslint-stylistic/releases) - [Commits](https://github.com/eslint-stylistic/eslint-stylistic/commits/v1.5.1/packages/eslint-plugin-js) --- updated-dependencies: - dependency-name: "@stylistic/eslint-plugin-js" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump eslint-plugin-n from 16.3.1 to 16.4.0 (#14017) * chore(deps-dev): bump eslint-plugin-n from 16.3.1 to 16.4.0 Bumps [eslint-plugin-n](https://github.com/eslint-community/eslint-plugin-n) from 16.3.1 to 16.4.0. - [Release notes](https://github.com/eslint-community/eslint-plugin-n/releases) - [Commits](https://github.com/eslint-community/eslint-plugin-n/compare/16.3.1...16.4.0) --- updated-dependencies: - dependency-name: eslint-plugin-n dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * fix(route): QuestMobile行业研究报告 (#14020) * fix(route): QuestMobile行业研究报告 * Update website/docs/routes/new-media.mdx --------- * fix(route): fix Yuque book route (#14022) * fix: fix Yuque book route * fix: sort switch conditions * fix: add cookieJar --------- * fix(route): picnob (#13986) * fix(route): picnob * fix(route): picnob. Use one browser session to do all http requests. * fix(route): picnob. Use puppeteer as a fallback option when a normal request returns a 403 error. * fix(route): picnob. Block unnecessary requests when using puppeteer. * fix(route): picnob. Adaptation of JSON responses when using puppeteer for http requests. * Update lib/v2/picnob/user.js --------- * fix(route/apnews): remove description (#14025) * chore(deps): bump github/codeql-action from 2 to 3 (#14026) Bumps [github/codeql-action](https://github.com/github/codeql-action) from 2 to 3. - [Release notes](https://github.com/github/codeql-action/releases) - [Changelog](https://github.com/github/codeql-action/blob/main/CHANGELOG.md) - [Commits](https://github.com/github/codeql-action/compare/v2...v3) --- updated-dependencies: - dependency-name: github/codeql-action dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/eslint from 8.44.8 to 8.44.9 (#14028) * chore(deps-dev): bump @types/eslint from 8.44.8 to 8.44.9 Bumps [@types/eslint](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/eslint) from 8.44.8 to 8.44.9. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/eslint) --- updated-dependencies: - dependency-name: "@types/eslint" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump @sentry/node from 7.86.0 to 7.87.0 (#14031) * chore(deps): bump @sentry/node from 7.86.0 to 7.87.0 Bumps [@sentry/node](https://github.com/getsentry/sentry-javascript) from 7.86.0 to 7.87.0. - [Release notes](https://github.com/getsentry/sentry-javascript/releases) - [Changelog](https://github.com/getsentry/sentry-javascript/blob/develop/CHANGELOG.md) - [Commits](https://github.com/getsentry/sentry-javascript/compare/7.86.0...7.87.0) --- updated-dependencies: - dependency-name: "@sentry/node" dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/react from 18.2.43 to 18.2.45 in /website (#14032) Bumps [@types/react](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/react) from 18.2.43 to 18.2.45. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/react) --- updated-dependencies: - dependency-name: "@types/react" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump puppeteer from 21.6.0 to 21.6.1 (#14029) * chore(deps): bump puppeteer from 21.6.0 to 21.6.1 Bumps [puppeteer](https://github.com/puppeteer/puppeteer) from 21.6.0 to 21.6.1. - [Release notes](https://github.com/puppeteer/puppeteer/releases) - [Changelog](https://github.com/puppeteer/puppeteer/blob/main/release-please-config.json) - [Commits](https://github.com/puppeteer/puppeteer/compare/puppeteer-v21.6.0...puppeteer-v21.6.1) --- updated-dependencies: - dependency-name: puppeteer dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @vercel/nft from 0.24.4 to 0.26.0 (#14030) * chore(deps-dev): bump @vercel/nft from 0.24.4 to 0.26.0 Bumps [@vercel/nft](https://github.com/vercel/nft) from 0.24.4 to 0.26.0. - [Release notes](https://github.com/vercel/nft/releases) - [Commits](https://github.com/vercel/nft/compare/0.24.4...0.26.0) --- updated-dependencies: - dependency-name: "@vercel/nft" dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump dawidd6/action-download-artifact from 2 to 3 (#14027) Bumps [dawidd6/action-download-artifact](https://github.com/dawidd6/action-download-artifact) from 2 to 3. - [Release notes](https://github.com/dawidd6/action-download-artifact/releases) - [Commits](https://github.com/dawidd6/action-download-artifact/compare/v2...v3) --- updated-dependencies: - dependency-name: dawidd6/action-download-artifact dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * docs: bring back remark formatter (#14040) * docs: format docs * docs: Fix URLs in website documentation * chore: bring back mdast formatter * docs: format docs * docs: remove heading id in jsx component * docs: fix heading level * chore: update remark formatter plugins * chore: Update dependabot ignore list * style: auto format * docs: fix table chore: update dependabot ignore * chore(deps): bump actions/upload-artifact from 3 to 4 (#14041) Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 3 to 4. - [Release notes](https://github.com/actions/upload-artifact/releases) - [Commits](https://github.com/actions/upload-artifact/compare/v3...v4) --- updated-dependencies: - dependency-name: actions/upload-artifact dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump eslint-plugin-yml from 1.10.0 to 1.11.0 (#14042) * chore(deps-dev): bump eslint-plugin-yml from 1.10.0 to 1.11.0 Bumps [eslint-plugin-yml](https://github.com/ota-meshi/eslint-plugin-yml) from 1.10.0 to 1.11.0. - [Release notes](https://github.com/ota-meshi/eslint-plugin-yml/releases) - [Changelog](https://github.com/ota-meshi/eslint-plugin-yml/blob/master/CHANGELOG.md) - [Commits](https://github.com/ota-meshi/eslint-plugin-yml/compare/v1.10.0...v1.11.0) --- updated-dependencies: - dependency-name: eslint-plugin-yml dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add 国家能源局发展规划司 (#14039) * style: auto format * chore(deps): bump @sentry/node from 7.87.0 to 7.88.0 (#14045) * chore(deps): bump @sentry/node from 7.87.0 to 7.88.0 Bumps [@sentry/node](https://github.com/getsentry/sentry-javascript) from 7.87.0 to 7.88.0. - [Release notes](https://github.com/getsentry/sentry-javascript/releases) - [Changelog](https://github.com/getsentry/sentry-javascript/blob/develop/CHANGELOG.md) - [Commits](https://github.com/getsentry/sentry-javascript/compare/7.87.0...7.88.0) --- updated-dependencies: - dependency-name: "@sentry/node" dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add 大连理工大学公共基础学院 RSS (#13982) (#14007) * Update index.js * fix(route): 中国政府网没有正确拼接相对地址路径 && 替换滚动新闻地址 * feat(route): add 大连理工大学公共基础学院 RSS (#13982) * 移除 pubDate 使用 new Date * 增加默认路由设置 * 完善公共基础学院说明文档 * 根据修改建议修改代码 * fix: sort maintainer --------- * style: auto format * feat(route): add fxiaoke.com blog (#14046) * feat(route): add fxiaoke.com blog * Update lib/v2/fxiaoke/radar.js * Update lib/v2/fxiaoke/radar.js * Update lib/v2/fxiaoke/radar.js * Update lib/v2/fxiaoke/radar.js * Update lib/v2/fxiaoke/radar.js * Update lib/v2/fxiaoke/radar.js * Update lib/v2/fxiaoke/crm.js * gen exact pubdate --------- * feat: title case following The Chicago Manual of Style (#14048) * feat(route): sspu (#14050) * fix(route): threads profile pic (#14061) * feat(route): add xhu people activities and answers (#14063) * style: auto format * feat(route): add 国家矿山安全监察局 (#14060) * style: auto format * fix(route): sehuatang append images in `.pattl` (#14055) * docs: fix xhu heading ids * docs: update badge * fix(route/reuters): Suppress full text fetch (#14035) * fix(route/deeplearning): The batch from deeplearning.ai (#14066) * fix(the-batch): The batch from deeplearning.ai * refactor: migrate to v2 --------- * fix(radar): 修复 xhu 用户动态匹配到自己的问题 (#14070) * docs: update badge * docs: update badge * feat: update github radars * fix: incorrect field name in UMS (#14073) * feat: remove notOperational routes - social media * feat(route): add tophub list 将榜单条目集合到一个列表中，可避免推送大量条目，更符合阅读习惯且有热度排序 (#14056) * feat(route): add fxiaoke.com blog * Update lib/v2/fxiaoke/radar.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/fxiaoke/radar.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/fxiaoke/radar.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/fxiaoke/radar.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/fxiaoke/radar.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/fxiaoke/radar.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/fxiaoke/crm.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * gen exact pubdate * feat(route): add tophub list * fix: guid * fix: use art to render rank --------- * feat(route): add artstation (#14075) * feat(route): add artstation * fix: update template * docs: add docs * feat: remove notOperational routes - new media * docs: fix badge text * chore: update stale config * chore(deps): bump pinyin-pro from 3.18.4 to 3.18.5 in /website (#14078) Bumps [pinyin-pro](https://github.com/zh-lx/pinyin-pro) from 3.18.4 to 3.18.5. - [Release notes](https://github.com/zh-lx/pinyin-pro/releases) - [Changelog](https://github.com/zh-lx/pinyin-pro/blob/main/CHANGELOG.md) - [Commits](https://github.com/zh-lx/pinyin-pro/compare/3.18.4...3.18.5) --- updated-dependencies: - dependency-name: pinyin-pro dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/react-dom in /website (#14079) Bumps [@types/react-dom](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/react-dom) from 18.2.17 to 18.2.18. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/react-dom) --- updated-dependencies: - dependency-name: "@types/react-dom" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump prism-react-renderer from 2.3.0 to 2.3.1 in /website (#14080) Bumps [prism-react-renderer](https://github.com/FormidableLabs/prism-react-renderer) from 2.3.0 to 2.3.1. - [Release notes](https://github.com/FormidableLabs/prism-react-renderer/releases) - [Commits](https://github.com/FormidableLabs/prism-react-renderer/compare/prism-react-renderer@2.3.0...prism-react-renderer@2.3.1) --- updated-dependencies: - dependency-name: prism-react-renderer dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/lint-staged from 13.2.2 to 13.3.0 (#14077) * chore(deps-dev): bump @types/lint-staged from 13.2.2 to 13.3.0 Bumps [@types/lint-staged](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/lint-staged) from 13.2.2 to 13.3.0. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/lint-staged) --- updated-dependencies: - dependency-name: "@types/lint-staged" dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump eslint from 8.55.0 to 8.56.0 (#14076) * chore(deps-dev): bump eslint from 8.55.0 to 8.56.0 Bumps [eslint](https://github.com/eslint/eslint) from 8.55.0 to 8.56.0. - [Release notes](https://github.com/eslint/eslint/releases) - [Changelog](https://github.com/eslint/eslint/blob/main/CHANGELOG.md) - [Commits](https://github.com/eslint/eslint/compare/v8.55.0...v8.56.0) --- updated-dependencies: - dependency-name: eslint dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): show weather info in qweather feed title (#14082) * feat(route): migrate jianshu to v2 (#14081) * feat(route): migrate jianshu to v2 * feat(docs): update jianshu docs * Refactor ProcessFeed function to improve code readability and performance --------- * feat: remove notOperational routes - traditional media * feat: remove notOperational routes - bbs * feat: remove notOperational routes - blog * feat(route): add 中国炼焦行业协会 (#14074) * feat(route): add 中国炼焦行业协会 * fix: remove subheadings for radar links * fix typo * feat: remove notOperational routes - programming * feat: remove notOperational routes - design * feat: remove notOperational routes - live * feat: remove notOperational routes - multimedia * chore(deps-dev): bump eslint-plugin-prettier from 5.0.1 to 5.1.0 (#14084) * chore(deps-dev): bump eslint-plugin-prettier from 5.0.1 to 5.1.0 Bumps [eslint-plugin-prettier](https://github.com/prettier/eslint-plugin-prettier) from 5.0.1 to 5.1.0. - [Release notes](https://github.com/prettier/eslint-plugin-prettier/releases) - [Changelog](https://github.com/prettier/eslint-plugin-prettier/blob/master/CHANGELOG.md) - [Commits](https://github.com/prettier/eslint-plugin-prettier/compare/v5.0.1...v5.1.0) --- updated-dependencies: - dependency-name: eslint-plugin-prettier dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add 中华人民共和国国家发展和改革委员会发展改革工作 (#14088) * feat: remove notOperational routes - picture * feat: hpoi * chore(deps-dev): bump eslint-plugin-n from 16.4.0 to 16.5.0 (#14091) * chore(deps-dev): bump eslint-plugin-n from 16.4.0 to 16.5.0 Bumps [eslint-plugin-n](https://github.com/eslint-community/eslint-plugin-n) from 16.4.0 to 16.5.0. - [Release notes](https://github.com/eslint-community/eslint-plugin-n/releases) - [Commits](https://github.com/eslint-community/eslint-plugin-n/compare/16.4.0...16.5.0) --- updated-dependencies: - dependency-name: eslint-plugin-n dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/eslint from 8.44.9 to 8.56.0 (#14092) * chore(deps-dev): bump @types/eslint from 8.44.9 to 8.56.0 Bumps [@types/eslint](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/eslint) from 8.44.9 to 8.56.0. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/eslint) --- updated-dependencies: - dependency-name: "@types/eslint" dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat: remove notOperational routes - anime * feat: remove notOperational routes - program-update * feat: remove notOperational routes - travel * feat(route): add mof (bond management) 中华人民共和国财政部-专题-政府债券管理 (#14094) * feat(route): add mof (bond management) * fix(router): re-order router config and add radar for mof * feat(router): radar param in route doc * fix(radar): add index source for mof * feat(radar): more source path for mof * Update website/docs/routes/government.mdx --------- * feat: remove notOperational routes - shopping * feat: remove notOperational routes - game * feat: remove notOperational routes - reading * feat: remove notOperational routes - study * feat: remove notOperational routes - journal * feat: remove notOperational routes - finance * feat: remove notOperational routes - other * chore(deps-dev): bump eslint-plugin-prettier from 5.1.0 to 5.1.1 (#14097) * chore(deps-dev): bump eslint-plugin-prettier from 5.1.0 to 5.1.1 Bumps [eslint-plugin-prettier](https://github.com/prettier/eslint-plugin-prettier) from 5.1.0 to 5.1.1. - [Release notes](https://github.com/prettier/eslint-plugin-prettier/releases) - [Changelog](https://github.com/prettier/eslint-plugin-prettier/blob/master/CHANGELOG.md) - [Commits](https://github.com/prettier/eslint-plugin-prettier/compare/v5.1.0...v5.1.1) --- updated-dependencies: - dependency-name: eslint-plugin-prettier dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * docs: fix path of zh configuration (#14100) * feat(route): add 中华人民共和国国家发展和改革委员会价格监测中心 (#14101) * chore(deps-dev): bump @types/supertest from 2.0.16 to 6.0.1 (#14104) * chore(deps-dev): bump @types/supertest from 2.0.16 to 6.0.1 Bumps [@types/supertest](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/supertest) from 2.0.16 to 6.0.1. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/supertest) --- updated-dependencies: - dependency-name: "@types/supertest" dependency-type: direct:development update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add consumer shopping-guide (#14105) * fix(route): dcfever (#14106) * chore(deps): bump @sentry/node from 7.88.0 to 7.89.0 (#14083) * chore(deps): bump @sentry/node from 7.88.0 to 7.89.0 Bumps [@sentry/node](https://github.com/getsentry/sentry-javascript) from 7.88.0 to 7.89.0. - [Release notes](https://github.com/getsentry/sentry-javascript/releases) - [Changelog](https://github.com/getsentry/sentry-javascript/blob/develop/CHANGELOG.md) - [Commits](https://github.com/getsentry/sentry-javascript/compare/7.88.0...7.89.0) --- updated-dependencies: - dependency-name: "@sentry/node" dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install * fix: replace deprecated `configureScope` in favor of `getCurrentScope()` ref: https://github.com/getsentry/sentry-javascript/blob/b27c2367acb312c4e9c2fd1aa2cdaf5b8cff1dad/MIGRATION.md --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add PKMer (#14103) * feat(route): PKMer * style: auto format * Update lib/v2/pkmer/recent.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/pkmer/radar.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update bbs.mdx * Update lib/v2/pkmer/recent.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> * fix(route): gamme (#14108) * feat(route): add 中国民用航空局公众留言 (#14109) * feat(route): 新增 bing 搜索; 迁移 bing 每日图片到 v2; 新增百度搜索; 迁移搜狗特色LOGO 到 v2 规范；添加搜狗搜索；添加 Google Search (#13936) * fix(route): 修复米游社公告栏 template 错误 * feat(route): 新增 bing 搜索 * docs: Update other.mdx * docs: fix docs * feat(route): 新增百度搜索 * fix(route): 修复 pubDate 解析错误 * fix(route): 优化百度搜索的缓存，减轻反爬问题 * feat(route): 新增 360 搜索 * feat(route): 迁移搜狗特色LOGO 到 v2 规范；添加搜狗搜索 * fix(route): 百度搜索增加图片 * feat(route): 新增 Google Search * fix(route): 修复百度搜索相关问题 * fix(route): 修复 Google 相关问题 * fix(route): 修复 360 搜索 * fix(route): 修复搜狗搜索 * fix(route): 修复 await 问题 * fix: 移除 google sites * fix(route): 修复缓存和过滤逻辑问题 * fix(route): 修复 360 搜索缺少 cookie 的问题 * fix(route): 修复 360 搜索 cookie 的问题 * feat(route): 移除 so.com 路由 * fix: merge conflict --------- * feat: add back blockbeats (#14113) * feat: add back blockbeats * fix: path * fix: adjust http log level (#14114) * feat: log redirect * fix: change puppeteer/proxy/redirect/got log level to `http` ref: https://github.com/winstonjs/winston#logging-levels (npm levels) * chore(deps): bump @tonyrl/rand-user-agent from 2.0.42 to 2.0.43 (#14117) * chore(deps): bump @tonyrl/rand-user-agent from 2.0.42 to 2.0.43 Bumps [@tonyrl/rand-user-agent](https://github.com/TonyRL/rand-user-agent) from 2.0.42 to 2.0.43. - [Release notes](https://github.com/TonyRL/rand-user-agent/releases) - [Commits](https://github.com/TonyRL/rand-user-agent/compare/v2.0.42...v2.0.43) --- updated-dependencies: - dependency-name: "@tonyrl/rand-user-agent" dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump eslint-plugin-prettier from 5.1.1 to 5.1.2 (#14116) * chore(deps-dev): bump eslint-plugin-prettier from 5.1.1 to 5.1.2 Bumps [eslint-plugin-prettier](https://github.com/prettier/eslint-plugin-prettier) from 5.1.1 to 5.1.2. - [Release notes](https://github.com/prettier/eslint-plugin-prettier/releases) - [Changelog](https://github.com/prettier/eslint-plugin-prettier/blob/master/CHANGELOG.md) - [Commits](https://github.com/prettier/eslint-plugin-prettier/compare/v5.1.1...v5.1.2) --- updated-dependencies: - dependency-name: eslint-plugin-prettier dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump @sentry/node from 7.89.0 to 7.91.0 (#14115) * chore(deps): bump @sentry/node from 7.89.0 to 7.91.0 Bumps [@sentry/node](https://github.com/getsentry/sentry-javascript) from 7.89.0 to 7.91.0. - [Release notes](https://github.com/getsentry/sentry-javascript/releases) - [Changelog](https://github.com/getsentry/sentry-javascript/blob/develop/CHANGELOG.md) - [Commits](https://github.com/getsentry/sentry-javascript/compare/7.89.0...7.91.0) --- updated-dependencies: - dependency-name: "@sentry/node" dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * fix(route): fix shiep/hhxy shiep/jsjxy (#14110) * fix(route): sort shiep config * fix(route): fix shiep/hhxy shiep/jsjxy * refactor: list processing in shiep/index.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * refactor: selector in shiep --------- * feat(route): add 界面新闻栏目 (#14121) * feat: set default itunes_explicit to false, close #14093 * PlayStation Monthly Games * feat(route): add new route for moj of gov.cn (#14122) * feat(route): add new route for moj of gov.cn * fix: typo in path Co-authored-by: Tony <TonyRL@users.noreply.github.com> --------- * chore(deps-dev): bump @types/supertest from 6.0.1 to 6.0.2 (#14126) * chore(deps-dev): bump @types/supertest from 6.0.1 to 6.0.2 Bumps [@types/supertest](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/supertest) from 6.0.1 to 6.0.2. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/supertest) --- updated-dependencies: - dependency-name: "@types/supertest" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): bluesky user posts (#14130) * chore(deps-dev): bump @types/react from 18.2.45 to 18.2.46 in /website (#14131) Bumps [@types/react](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/react) from 18.2.45 to 18.2.46. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/react) --- updated-dependencies: - dependency-name: "@types/react" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add Cool Papers (#14129) * feat(route): add Cool Papers * fix typo * fix: remove kimi chat content * docs: fix typo --------- * style: auto format * chore(deps-dev): bump eslint-plugin-n from 16.5.0 to 16.6.0 (#14140) * chore(deps-dev): bump eslint-plugin-n from 16.5.0 to 16.6.0 Bumps [eslint-plugin-n](https://github.com/eslint-community/eslint-plugin-n) from 16.5.0 to 16.6.0. - [Release notes](https://github.com/eslint-community/eslint-plugin-n/releases) - [Commits](https://github.com/eslint-community/eslint-plugin-n/compare/16.5.0...16.6.0) --- updated-dependencies: - dependency-name: eslint-plugin-n dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump koa from 2.14.2 to 2.15.0 (#14141) * chore(deps): bump koa from 2.14.2 to 2.15.0 Bumps [koa](https://github.com/koajs/koa) from 2.14.2 to 2.15.0. - [Changelog](https://github.com/koajs/koa/blob/2.15.0/History.md) - [Commits](https://github.com/koajs/koa/compare/2.14.2...2.15.0) --- updated-dependencies: - dependency-name: koa dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump clsx from 2.0.0 to 2.1.0 in /website (#14142) Bumps [clsx](https://github.com/lukeed/clsx) from 2.0.0 to 2.1.0. - [Release notes](https://github.com/lukeed/clsx/releases) - [Commits](https://github.com/lukeed/clsx/compare/v2.0.0...v2.1.0) --- updated-dependencies: - dependency-name: clsx dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat: add air quality content in qweather 3days forecast feed (#14136) * feat: add air quality content in qweather 3days report feed * chore: title enhancement * fix: resolve no needed lines * fix: add guard for api key config * docs: Update InstanceList.tsx - add an instance (#14143) add https://rsshub.rss.tips * feat(route): 三联生活周刊 (#14127) * feat(route): 三联生活周刊 * fix: namespace and data acquirement * fix: get article list by api * Update lib/v2/lifeweek/channel.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/lifeweek/channel.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/lifeweek/tag.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * fix: rss url * refactor: getRssItem function * Update lib/v2/lifeweek/utils.js --------- Co-authored-by: Changren Wang <changren.wcr@alibaba-inc.com> * feat(route): add 中国的中古 (#14139) * feat(route): add 中国的中古 * Update lib/v2/medieval-china/post.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * Update lib/v2/medieval-china/maintainer.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * fix(route): fix desc of 中国的中古 * fix(route): fix data query of 中国的中古 * Update lib/v2/medieval-china/post.js Co-authored-by: Tony <TonyRL@users.noreply.github.com> * style: auto format * fix(radar): add target --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> * feat(route): id param for shmeea (#14145) * feat(route): id param for shmeea * fix(route): fix radar docs url in shmeea/self-study * fix(route): id?=08000 && block requests to binary files && fix code style --------- * feat(route): 添加 dm_img_list 字段防止被 Bilibili 识别 (#14128) * feat(route): add dm_img_list parameter Signed-off-by: NightSpaceC <NightSpaceC@outlook.com> * feat(route): readd video-all. It is still usable. * fix(route): use camelCase * feat(route): generate data of dm_img_list --------- Signed-off-by: NightSpaceC <NightSpaceC@outlook.com> * style: auto format * feat(route): add 国家气候中心最新监测 (#14151) * feat(route): add 国家气候中心最新监测 * fix typo * feat(route): add 上海第二工业大学体育部 (#14149) * feat(route): add 上海第二工业大学体育部 * fix docs * fix: block requests to binary files * fix(route): douban recommended two-digit month (#14153) * chore(deps): bump @tonyrl/rand-user-agent from 2.0.43 to 2.0.44 (#14156) * chore(deps): bump @tonyrl/rand-user-agent from 2.0.43 to 2.0.44 Bumps [@tonyrl/rand-user-agent](https://github.com/TonyRL/rand-user-agent) from 2.0.43 to 2.0.44. - [Release notes](https://github.com/TonyRL/rand-user-agent/releases) - [Commits](https://github.com/TonyRL/rand-user-agent/compare/v2.0.43...v2.0.44) --- updated-dependencies: - dependency-name: "@tonyrl/rand-user-agent" dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump chrono-node from 2.7.3 to 2.7.4 (#14155) * chore(deps): bump chrono-node from 2.7.3 to 2.7.4 Bumps [chrono-node](https://github.com/wanasit/chrono) from 2.7.3 to 2.7.4. - [Release notes](https://github.com/wanasit/chrono/releases) - [Commits](https://github.com/wanasit/chrono/compare/v2.7.3...v2.7.4) --- updated-dependencies: - dependency-name: chrono-node dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add OTOBANANA (#14160) * feat(route): add OTOBANANA * fix: live guid * feat: utgd add premium identification (#14163) * fix(route): 界面新闻重复文章视频 (#14162) * fix(route): 界面新闻重复文章视频 * fix: improve url sanitization * feat(route): backlinko (#14164) * feat(route): backlinko * Refactor blog.js to destructure nested properties * chore(deps): bump pinyin-pro from 3.18.5 to 3.18.6 in /website (#14167) Bumps [pinyin-pro](https://github.com/zh-lx/pinyin-pro) from 3.18.5 to 3.18.6. - [Release notes](https://github.com/zh-lx/pinyin-pro/releases) - [Changelog](https://github.com/zh-lx/pinyin-pro/blob/main/CHANGELOG.md) - [Commits](https://github.com/zh-lx/pinyin-pro/compare/3.18.5...3.18.6) --- updated-dependencies: - dependency-name: pinyin-pro dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @stylistic/eslint-plugin-js from 1.5.1 to 1.5.3 (#14165) * chore(deps-dev): bump @stylistic/eslint-plugin-js from 1.5.1 to 1.5.3 Bumps [@stylistic/eslint-plugin-js](https://github.com/eslint-stylistic/eslint-stylistic/tree/HEAD/packages/eslint-plugin-js) from 1.5.1 to 1.5.3. - [Release notes](https://github.com/eslint-stylistic/eslint-stylistic/releases) - [Commits](https://github.com/eslint-stylistic/eslint-stylistic/commits/v1.5.3/packages/eslint-plugin-js) --- updated-dependencies: - dependency-name: "@stylistic/eslint-plugin-js" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump eslint-plugin-n from 16.6.0 to 16.6.1 (#14166) * chore(deps-dev): bump eslint-plugin-n from 16.6.0 to 16.6.1 Bumps [eslint-plugin-n](https://github.com/eslint-community/eslint-plugin-n) from 16.6.0 to 16.6.1. - [Release notes](https://github.com/eslint-community/eslint-plugin-n/releases) - [Commits](https://github.com/eslint-community/eslint-plugin-n/compare/16.6.0...16.6.1) --- updated-dependencies: - dependency-name: eslint-plugin-n dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * fix(route): 第一财经DT财经 (#14171) * chore(deps-dev): bump @types/imapflow from 1.0.16 to 1.0.17 (#14172) * chore(deps-dev): bump @types/imapflow from 1.0.16 to 1.0.17 Bumps [@types/imapflow](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/imapflow) from 1.0.16 to 1.0.17. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/imapflow) --- updated-dependencies: - dependency-name: "@types/imapflow" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/eslint from 8.56.0 to 8.56.1 (#14173) * chore(deps-dev): bump @types/eslint from 8.56.0 to 8.56.1 Bumps [@types/eslint](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/eslint) from 8.56.0 to 8.56.1. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/eslint) --- updated-dependencies: - dependency-name: "@types/eslint" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @vercel/nft from 0.26.0 to 0.26.2 (#14174) * chore(deps-dev): bump @vercel/nft from 0.26.0 to 0.26.2 Bumps [@vercel/nft](https://github.com/vercel/nft) from 0.26.0 to 0.26.2. - [Release notes](https://github.com/vercel/nft/releases) - [Commits](https://github.com/vercel/nft/compare/0.26.0...0.26.2) --- updated-dependencies: - dependency-name: "@vercel/nft" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat: recover /ft/myft, close #14096 * fix(route): 第一财经DT财经报告附件 (#14176) * feat(route): recover gofans (#14183) * chore(deps): bump mailparser from 3.6.5 to 3.6.6 (#14184) * chore(deps): bump mailparser from 3.6.5 to 3.6.6 Bumps [mailparser](https://github.com/nodemailer/mailparser) from 3.6.5 to 3.6.6. - [Release notes](https://github.com/nodemailer/mailparser/releases) - [Changelog](https://github.com/nodemailer/mailparser/blob/master/CHANGELOG.md) - [Commits](https://github.com/nodemailer/mailparser/compare/v3.6.5...v3.6.6) --- updated-dependencies: - dependency-name: mailparser dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump puppeteer from 21.6.1 to 21.7.0 (#14185) * chore(deps): bump puppeteer from 21.6.1 to 21.7.0 Bumps [puppeteer](https://github.com/puppeteer/puppeteer) from 21.6.1 to 21.7.0. - [Release notes](https://github.com/puppeteer/puppeteer/releases) - [Changelog](https://github.com/puppeteer/puppeteer/blob/main/release-please-config.json) - [Commits](https://github.com/puppeteer/puppeteer/compare/puppeteer-v21.6.1...puppeteer-v21.7.0) --- updated-dependencies: - dependency-name: puppeteer dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump @sentry/node from 7.91.0 to 7.92.0 (#14186) * chore(deps): bump @sentry/node from 7.91.0 to 7.92.0 Bumps [@sentry/node](https://github.com/getsentry/sentry-javascript) from 7.91.0 to 7.92.0. - [Release notes](https://github.com/getsentry/sentry-javascript/releases) - [Changelog](https://github.com/getsentry/sentry-javascript/blob/develop/CHANGELOG.md) - [Commits](https://github.com/getsentry/sentry-javascript/compare/7.91.0...7.92.0) --- updated-dependencies: - dependency-name: "@sentry/node" dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * fix: domp4 supports secondary address (#14181) (#14189) * fix: domp4 supports secondary address * fix: domp4 remove invalid domain * feat(route): 调整生成鼠标路径的参数，在配置中预置路径 (#14179) * fix(route): adjust the parameter to generate path * feat(route): use the path from configure * feat(docs): add the usage of BILIBILI_DM_IMG_LIST * style: auto format * feat: New Router for liveuamap (#14175) * Added new route for liveuamap * Fix unsafe domain and the 3rd level domain defaulting * docs: fix heading --------- * chore(deps): bump googleapis from 129.0.0 to 130.0.0 (#14193) * chore(deps): bump googleapis from 129.0.0 to 130.0.0 Bumps [googleapis](https://github.com/googleapis/google-api-nodejs-client) from 129.0.0 to 130.0.0. - [Release notes](https://github.com/googleapis/google-api-nodejs-client/releases) - [Changelog](https://github.com/googleapis/google-api-nodejs-client/blob/main/release-please-config.json) - [Commits](https://github.com/googleapis/google-api-nodejs-client/compare/googleapis-v129.0.0...googleapis-v130.0.0) --- updated-dependencies: - dependency-name: googleapis dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump jsdom from 23.0.1 to 23.1.0 (#14194) * chore(deps): bump jsdom from 23.0.1 to 23.1.0 Bumps [jsdom](https://github.com/jsdom/jsdom) from 23.0.1 to 23.1.0. - [Release notes](https://github.com/jsdom/jsdom/releases) - [Changelog](https://github.com/jsdom/jsdom/blob/main/Changelog.md) - [Commits](https://github.com/jsdom/jsdom/compare/23.0.1...23.1.0) --- updated-dependencies: - dependency-name: jsdom dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump the docusaurus group in /website with 7 updates (#14192) Bumps the docusaurus group in /website with 7 updates: | Package | From | To | | --- | --- | --- | | [@docusaurus/core](https://github.com/facebook/docusaurus/tree/HEAD/packages/docusaurus) | `3.0.1` | `3.1.0` | | [@docusaurus/plugin-client-redirects](https://github.com/facebook/docusaurus/tree/HEAD/packages/docusaurus-plugin-client-redirects) | `3.0.1` | `3.1.0` | | [@docusaurus/plugin-pwa](https://github.com/facebook/docusaurus/tree/HEAD/packages/docusaurus-plugin-pwa) | `3.0.1` | `3.1.0` | | [@docusaurus/preset-classic](https://github.com/facebook/docusaurus/tree/HEAD/packages/docusaurus-preset-classic) | `3.0.1` | `3.1.0` | | [@docusaurus/module-type-aliases](https://github.com/facebook/docusaurus/tree/HEAD/packages/docusaurus-module-type-aliases) | `3.0.1` | `3.1.0` | | [@docusaurus/tsconfig](https://github.com/facebook/docusaurus/tree/HEAD/packages/docusaurus-tsconfig) | `3.0.1` | `3.1.0` | | [@docusaurus/types](https://github.com/facebook/docusaurus/tree/HEAD/packages/docusaurus-types) | `3.0.1` | `3.1.0` | Updates `@docusaurus/core` from 3.0.1 to 3.1.0 - [Release notes](https://github.com/facebook/docusaurus/releases) - [Changelog](https://github.com/facebook/docusaurus/blob/main/CHANGELOG.md) - [Commits](https://github.com/facebook/docusaurus/commits/v3.1.0/packages/docusaurus) Updates `@docusaurus/plugin-client-redirects` from 3.0.1 to 3.1.0 - [Release notes](https://github.com/facebook/docusaurus/releases) - [Changelog](https://github.com/facebook/docusaurus/blob/main/CHANGELOG.md) - [Commits](https://github.com/facebook/docusaurus/commits/v3.1.0/packages/docusaurus-plugin-client-redirects) Updates `@docusaurus/plugin-pwa` from 3.0.1 to 3.1.0 - [Release notes](https://github.com/facebook/docusaurus/releases) - [Changelog](https://github.com/facebook/docusaurus/blob/main/CHANGELOG.md) - [Commits](https://github.com/facebook/docusaurus/commits/v3.1.0/packages/docusaurus-plugin-pwa) Updates `@docusaurus/preset-classic` from 3.0.1 to 3.1.0 - [Release notes](https://github.com/facebook/docusaurus/releases) - [Changelog](https://github.com/facebook/docusaurus/blob/main/CHANGELOG.md) - [Commits](https://github.com/facebook/docusaurus/commits/v3.1.0/packages/docusaurus-preset-classic) Updates `@docusaurus/module-type-aliases` from 3.0.1 to 3.1.0 - [Release notes](https://github.com/facebook/docusaurus/releases) - [Changelog](https://github.com/facebook/docusaurus/blob/main/CHANGELOG.md) - [Commits](https://github.com/facebook/docusaurus/commits/v3.1.0/packages/docusaurus-module-type-aliases) Updates `@docusaurus/tsconfig` from 3.0.1 to 3.1.0 - [Release notes](https://github.com/facebook/docusaurus/releases) - [Changelog](https://github.com/facebook/docusaurus/blob/main/CHANGELOG.md) - [Commits](https://github.com/facebook/docusaurus/commits/v3.1.0/packages/docusaurus-tsconfig) Updates `@docusaurus/types` from 3.0.1 to 3.1.0 - [Release notes](https://github.com/facebook/docusaurus/releases) - [Changelog](https://github.com/facebook/docusaurus/blob/main/CHANGELOG.md) - [Commits](https://github.com/facebook/docusaurus/commits/v3.1.0/packages/docusaurus-types) --- updated-dependencies: - dependency-name: "@docusaurus/core" dependency-type: direct:production update-type: version-update:semver-minor dependency-group: docusaurus - dependency-name: "@docusaurus/plugin-client-redirects" dependency-type: direct:production update-type: version-update:semver-minor dependency-group: docusaurus - dependency-name: "@docusaurus/plugin-pwa" dependency-type: direct:production update-type: version-update:semver-minor dependency-group: docusaurus - dependency-name: "@docusaurus/preset-classic" dependency-type: direct:production update-type: version-update:semver-minor dependency-group: docusaurus - dependency-name: "@docusaurus/module-type-aliases" dependency-type: direct:development update-type: version-update:semver-minor dependency-group: docusaurus - dependency-name: "@docusaurus/tsconfig" dependency-type: direct:development update-type: version-update:semver-minor dependency-group: docusaurus - dependency-name: "@docusaurus/types" dependency-type: direct:development update-type: version-update:semver-minor dependency-group: docusaurus ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * fix(route): Tradingview Blog (#14191) * fix(route): Tradingview Blog * fix: use tiny-async-pool * fix(route): fix shiep/jsjxy shiep/jwc (#14196) * docs: change html example to use items variable instead of item (#14205) * change html example to use items variable instead of item The existing example for HTML retrieval uses 'item' variable on item retrieval, but the final rss output uses the 'items' variable. This results in undefined variable for anyone who directly uses the example code. * docs: fix cn docs too --------- * feat(route): add 中华全国专利代理师协会 (#14197) * fix: zhihu timeline (#14169) * fix zhihu timeline * deal with content in an array * adopt content_html if exists * fix(route): 处理大麦网查询结果为空的情况 (#14203) * fix(route): 处理大麦网查询结果为空的情况 * refactor: migrate to v2 --------- * chore(deps): bump @tonyrl/rand-user-agent from 2.0.44 to 2.0.45 (#14208) * chore(deps): bump @tonyrl/rand-user-agent from 2.0.44 to 2.0.45 Bumps [@tonyrl/rand-user-agent](https://github.com/TonyRL/rand-user-agent) from 2.0.44 to 2.0.45. - [Release notes](https://github.com/TonyRL/rand-user-agent/releases) - [Commits](https://github.com/TonyRL/rand-user-agent/compare/v2.0.44...v2.0.45) --- updated-dependencies: - dependency-name: "@tonyrl/rand-user-agent" dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps): bump jsdom from 23.1.0 to 23.2.0 (#14209) * chore(deps): bump jsdom from 23.1.0 to 23.2.0 Bumps [jsdom](https://github.com/jsdom/jsdom) from 23.1.0 to 23.2.0. - [Release notes](https://github.com/jsdom/jsdom/releases) - [Changelog](https://github.com/jsdom/jsdom/blob/main/Changelog.md) - [Commits](https://github.com/jsdom/jsdom/compare/23.1.0...23.2.0) --- updated-dependencies: - dependency-name: jsdom dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/react from 18.2.46 to 18.2.47 in /website (#14210) Bumps [@types/react](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/react) from 18.2.46 to 18.2.47. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/react) --- updated-dependencies: - dependency-name: "@types/react" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): Add Onet (#14200) * feat(route): Add Onet * use arrow function in `router.js` * Update lib/v2/onet/templates/article.art * Update lib/v2/onet/templates/image.art * Update lib/v2/onet/maintainer.js * Update website/docs/routes/new-media.mdx --------- * fix(route): saraba1st digest image (#14206) * fix saraba1st digest image * add missing semicolon * chore(deps-dev): bump @types/koa from 2.13.12 to 2.14.0 (#14215) * chore(deps-dev): bump @types/koa from 2.13.12 to 2.14.0 Bumps [@types/koa](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/koa) from 2.13.12 to 2.14.0. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/koa) --- updated-dependencies: - dependency-name: "@types/koa" dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * feat(route): add huggingface zh blog (#14211) * feat(route): add huggingface zh blog * refactor: update * feat(route): Add ekantipur.com (Nepal) (#14207) * feat(route): Add ekantipur remove unused deps * feat(route): Add ekantipur radar * removed undefined field * updated maintainer.js to use optional field character - ? * updated radar.js with full name --------- * style: auto format * feat(route): HoYoLAB (#14146) * hoyolab * 修改分页参数 * 统一名称 * 替换limit参数 * 参数默认值问题 * 参数默认值问题 * feat(route): trending papers on arXiv from trendingpapers (#14182) * style: auto format * chore(deps): bump @sentry/node from 7.92.0 to 7.93.0 (#14220) * chore(deps): bump @sentry/node from 7.92.0 to 7.93.0 Bumps [@sentry/node](https://github.com/getsentry/sentry-javascript) from 7.92.0 to 7.93.0. - [Release notes](https://github.com/getsentry/sentry-javascript/releases) - [Changelog](https://github.com/getsentry/sentry-javascript/blob/develop/CHANGELOG.md) - [Commits](https://github.com/getsentry/sentry-javascript/compare/7.92.0...7.93.0) --- updated-dependencies: - dependency-name: "@sentry/node" dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump eslint-plugin-n from 16.6.1 to 16.6.2 (#14216) * chore(deps-dev): bump eslint-plugin-n from 16.6.1 to 16.6.2 Bumps [eslint-plugin-n](https://github.com/eslint-community/eslint-plugin-n) from 16.6.1 to 16.6.2. - [Release notes](https://github.com/eslint-community/eslint-plugin-n/releases) - [Commits](https://github.com/eslint-community/eslint-plugin-n/compare/16.6.1...16.6.2) --- updated-dependencies: - dependency-name: eslint-plugin-n dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump eslint-plugin-prettier from 5.1.2 to 5.1.3 (#14221) * chore(deps-dev): bump eslint-plugin-prettier from 5.1.2 to 5.1.3 Bumps [eslint-plugin-prettier](https://github.com/prettier/eslint-plugin-prettier) from 5.1.2 to 5.1.3. - [Release notes](https://github.com/prettier/eslint-plugin-prettier/releases) - [Changelog](https://github.com/prettier/eslint-plugin-prettier/blob/master/CHANGELOG.md) - [Commits](https://github.com/prettier/eslint-plugin-prettier/compare/v5.1.2...v5.1.3) --- updated-dependencies: - dependency-name: eslint-plugin-prettier dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * fix: 修正 cyzone 返回link 为空 (#14224) * fix(egsea): pubDate Invalid Date the https://rsshub.app/egsea/flash response error pubDate <pubDate>Invalid Date</pubDate> * refactor: migrate to v2 * Refactor link generation in util.js --------- * fix(route): 国家外汇管理局业务咨询 & 投诉建议链接 (#14226) * chore(deps): bump pinyin-pro from 3.18.6 to 3.19.0 in /website (#14222) Bumps [pinyin-pro](https://github.com/zh-lx/pinyin-pro) from 3.18.6 to 3.19.0. - [Release notes](https://github.com/zh-lx/pinyin-pro/releases) - [Changelog](https://github.com/zh-lx/pinyin-pro/blob/main/CHANGELOG.md) - [Commits](https://github.com/zh-lx/pinyin-pro/commits) --- updated-dependencies: - dependency-name: pinyin-pro dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * chore(deps-dev): bump @types/eslint from 8.56.1 to 8.56.2 (#14228) * chore(deps-dev): bump @types/eslint from 8.56.1 to 8.56.2 Bumps [@types/eslint](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/eslint) from 8.56.1 to 8.56.2. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/eslint) --- updated-dependencies: - dependency-name: "@types/eslint" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * docs: fix maintainer * feat(route): add BT之家1LOU站 (#14219) * route: BT之家1LOU * 1. using new routers. 2. If path is empty, visit correct main page of website. 3. roll back deleted doc by mistake. * Update lib/v2/1lou/radar.js * Update website/docs/routes/multimedia.mdx * Update website/docs/routes/multimedia.mdx --------- * style: auto format * chore(deps-dev): bump prettier from 3.1.1 to 3.2.1 (#14230) * chore(deps-dev): bump prettier from 3.1.1 to 3.2.1 Bumps [prettier](https://github.com/prettier/prettier) from 3.1.1 to 3.2.1. - [Release notes](https://github.com/prettier/prettier/releases) - [Changelog](https://github.com/prettier/prettier/blob/main/CHANGELOG.md) - [Commits](https://github.com/prettier/prettier/compare/3.1.1...3.2.1) --- updated-dependencies: - dependency-name: prettier dependency-type: direct:development update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> * chore: fix pnpm install --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * style: auto format * feat(proxy): add PAC script support (#14218) * feat(proxy): add PAC script support * fix pnpm fail * fix coverage * fix coverage final * update docs * feat(ps): use art * feat(route): add TradingView Desktop releases and release notes (#14234) * fix: pornhub pubDate (#14232) * fix(route): fix twitch (#14238) * fix(route): fix twitch * fix docs * feat(route): add 苏州市发展和改革委员会 (#14214) * feat(route): add 苏州市发展和改革委员会 * docs: remove duplicated heading * refactor: migrate to v2 * fix: suzhou docs * fix: news --------- * fix(route): tencent author (#14241) * fix: luogu route parse error (#14170) * fix route parse error * use parse-date instead of Date * optimize decode processes * fix typo --------- * fix(route): hoyolab (#14242) * feat: support CIDR IP ranges in allowlist (#14243) * docs: Update InstanceList.tsx (#14244) Add instance hosted by Kai. * feat(route): recover kuwaitlocal agirls qianp taiwannews jiaoliudao (#14247) * fix: recover kuwaitlocal * fix: recover agirls * fix: recover qianp * fix: recover taiwannews * fix: recover hket * fix: recover jiaoliudao * fix: qianp * fix: deepscan issue * feat(route): zhihu xhu posts (#14246) * feat: recover shuiguopai (#14248) * chore(deps-dev): bump @types/react from 18.2.47 to 18.2.48 in /website (#14255) Bumps [@types/react](https://github.com/DefinitelyTyped/DefinitelyTyped/tree/HEAD/types/react) from 18.2.47 to 18.2.48. - [Release notes](https://github.com/DefinitelyTyped/DefinitelyTyped/releases) - [Commits](https://github.com/DefinitelyTyped/DefinitelyTyped/commits/HEAD/types/react) --- updated-dependencies: - dependency-name: "@types/react" dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.nor…

Force1ess · 2024-02-29T15:24:01Z

@nczitzk
Hello, thank you for your PR. I appreciate your contribution.
However, it seems that one of the rules in question is no longer valid.
Could you please help to address this?
Thanks

nczitzk · 2024-03-01T23:28:14Z

However, it seems that one of the rules in question is no longer valid.

I am working on it. 😊

Force1ess · 2024-03-02T02:52:48Z

However, it seems that one of the rules in question is no longer valid.但是，似乎其中一条规则不再有效。

I am working on it. 😊我正在努力。😊

I am very happy to receive your reply.
Thank you for your selfless dedication.

nczitzk added 2 commits December 27, 2023 12:10

feat(route): add Cool Papers

f06663c

fix typo

58b6af9

github-actions bot added Route: v2 v2 route related Auto: Route Test Complete Auto route test has finished on given PR labels Dec 28, 2023

TonyRL reviewed Dec 28, 2023

View reviewed changes

fix: remove kimi chat content

4b7038c

TonyRL reviewed Dec 29, 2023

View reviewed changes

website/docs/routes/journal.mdx Outdated Show resolved Hide resolved

docs: fix typo

df1d983

Co-authored-by: Tony <TonyRL@users.noreply.github.com>

TonyRL merged commit e7f20b4 into DIYgod:master Dec 29, 2023
34 checks passed

nczitzk deleted the feature/papers branch December 29, 2023 14:36

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(route): add Cool Papers #14129

feat(route): add Cool Papers #14129

nczitzk commented Dec 28, 2023

github-actions bot commented Dec 28, 2023

TonyRL Dec 28, 2023

TonyRL Dec 28, 2023

TonyRL Dec 28, 2023

nczitzk Dec 29, 2023

github-actions bot commented Dec 29, 2023

github-actions bot commented Dec 29, 2023

Force1ess commented Feb 29, 2024

nczitzk commented Mar 1, 2024

Force1ess commented Mar 2, 2024

feat(route): add Cool Papers #14129

feat(route): add Cool Papers #14129

Conversation

nczitzk commented Dec 28, 2023

Involved Issue / 该 PR 相关 Issue

Example for the Proposed Route(s) / 路由地址示例

New RSS Route Checklist / 新 RSS 路由检查表

Note / 说明

github-actions bot commented Dec 28, 2023

TonyRL Dec 28, 2023

Choose a reason for hiding this comment

TonyRL Dec 28, 2023

Choose a reason for hiding this comment

TonyRL Dec 28, 2023

Choose a reason for hiding this comment

nczitzk Dec 29, 2023

Choose a reason for hiding this comment

github-actions bot commented Dec 29, 2023

github-actions bot commented Dec 29, 2023

Force1ess commented Feb 29, 2024

nczitzk commented Mar 1, 2024

Force1ess commented Mar 2, 2024