NeurIPS 2022accept38 citations
Second Thoughts are Best: Learning to Re-Align With Human Values from Text Edits
Ruibo Liu, Chenyan Jia, Ge Zhang, Ziyu Zhuang, Tony X Liu, Soroush Vosoughi
Abstract
We present Second Thoughts, a new learning paradigm that enables language models (LMs) to re-align with human values. By modeling the chain-of-edits between value-unaligned and value-aligned text, with LM fine-tuning and additional refinement through reinforcement learning, Second Thoughts not only achieves superior performance in three value alignment benchmark datasets but also shows strong human-value transfer learning ability in few-shot scenarios. The generated editing steps also offer better interpretability and ease for interactive error correction. Extensive human evaluations further confirm its effectiveness.
human valuesai safetyalignmentsocial impacthuman-AI interaction
BibTeX
@inproceedings{
liu2022second,
title={Second Thoughts are Best: Learning to Re-Align With Human Values from Text Edits},
author={Ruibo Liu and Chenyan Jia and Ge Zhang and Ziyu Zhuang and Tony X Liu and Soroush Vosoughi},
booktitle={Advances in Neural Information Processing Systems},
editor={Alice H. Oh and Alekh Agarwal and Danielle Belgrave and Kyunghyun Cho},
year={2022},
url={https://openreview.net/forum?id=u6OfmaGIya1}
}