Consistency Is Not Validity: Self-Rewarded Moral Argumentation Makes Language Models Decisive but Pliable
arXiv:2610.30349v1 Announce Type: new Abstract: LLMs are increasingly consulted for moral judgment. We aim to improve performance on such tasks by rewarding good arguments, on the premise that arguing well increases the likelihood of judging well. We train a reinforcement learning policy initialize…
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-10-24 04:00 · arXiv cs.CL
Consistency Is Not Validity: Self-Rewarded Moral Argumentation Makes Language Models Decisive but Pliable