ClarifyDelphi: Reinforced Clarification Questions with Defeasibility Rewards for Social and Moral Situations

Valentina Pyatkin, Jena D. Hwang, Vivek Srikumar, Ximing Lu, Liwei Jiang, Yejin Choi, Chandra Bhagavatula


Abstract
Context is everything, even in commonsense moral reasoning. Changing contexts can flip the moral judgment of an action; Lying to a friend is wrong in general, but may be morally acceptable if it is intended to protect their life. We present ClarifyDelphi, an interactive system that learns to ask clarification questions (e.g., why did you lie to your friend?) in order to elicit additional salient contexts of a social or moral situation. We posit that questions whose potential answers lead to diverging moral judgments are the most informative. Thus, we propose a reinforcement learning framework with a defeasibility reward that aims to maximize the divergence between moral judgments of hypothetical answers to a question. Human evaluation demonstrates that our system generates more relevant, informative and defeasible questions compared to competitive baselines. Our work is ultimately inspired by studies in cognitive science that have investigated the flexibility in moral cognition (i.e., the diverse contexts in which moral rules can be bent), and we hope that research in this direction can assist both cognitive and computational investigations of moral judgments.
Anthology ID:
2023.acl-long.630
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
11253–11271
Language:
URL:
https://aclanthology.org/2023.acl-long.630
DOI:
10.18653/v1/2023.acl-long.630
Bibkey:
Cite (ACL):
Valentina Pyatkin, Jena D. Hwang, Vivek Srikumar, Ximing Lu, Liwei Jiang, Yejin Choi, and Chandra Bhagavatula. 2023. ClarifyDelphi: Reinforced Clarification Questions with Defeasibility Rewards for Social and Moral Situations. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11253–11271, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
ClarifyDelphi: Reinforced Clarification Questions with Defeasibility Rewards for Social and Moral Situations (Pyatkin et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-long.630.pdf
Video:
 https://aclanthology.org/2023.acl-long.630.mp4