Xingwu Chen, Zhanqiu Zhang, Yiwen Guo, Difan Zou

Breaking Contextual Inertia: Reinforcement Learning with Single-Turn Anchors for Stable Multi-Turn Interaction

arXiv:2603.04783v1 Announce Type: new Abstract: While LLMs demonstrate strong reasoning capabilities when provided with full information in a single turn, they exhibit substantial vulnerability in …