Reframing Human-AI Collaboration for Generating Free-Text Explanations

Repository for the NAACL 2022 paper "Reframing Human-AI Collaboration for Generating Free-Text Explanations" by Sarah Wiegreffe, Jack Hessel, Swabha Swayamdipta, Mark Riedl, and Yejin Choi.

When using this repository, please cite:

@inproceedings{wiegreffe2021reframing,
  title={Reframing Human-AI Collaboration for Generating Free-Text Explanations},
  author={Wiegreffe, Sarah and Hessel, Jack and Swayamdipta, Swabha and Riedl, Mark and Choi, Yejin},
  year={2022},
  booktitle={Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies},
  month=jul,
  publisher={Association for Computational Linguistics},
  address="Online and Seattle, USA",
  url={https://arxiv.org/abs/2112.08674/}
}

Pretrained models

See the pretrained_models/ directory for checkpoints and demo code.

Data

See the data/ directory for GPT-3 generations and crowdsourced acceptability annotations. We obtained approval from OpenAI to release the GPT-3-generated explanations.

GPT-3 Prompting Scripts

See the gpt3_prompting_scripts/ directory.

Amazon Mechanical Turk templates

See the mechanicalTurk_templates/ directory.