AI RESEARCH

Mitigating Lost in Multi-turn Conversation via Curriculum RL with Verifiable Accuracy and Abstention Rewards

arXiv CS.LG

ArXi:2510.18731v2 Announce Type: replace-cross Large Language Models nstrate strong capabilities in single-turn instruction following but suffer from Lost-in-Conversation (LiC), a degradation in performance as information is revealed progressively in multi-turn settings.