TRUSTMEM：为具有长期记忆的 LLM 代理学习可信赖的记忆整合

arXiv:2606.25161v1 公告类型：新 摘要：大型语言模型 (LLM) 代理依靠长期记忆来支持有限上下文窗口之外的扩展交互和个性化帮助。 Existing memory agents actively update external memory through generated write, revise, and delete operations, but these updates may omit important information, corrupt existing memory, or introduce unsupported hallucinated content.一旦存储，此类错误就会成为持久的系统状态故障，可能影响未来的推理和生成。在本文中，我们提出了 TrustMem，一个旨在提高内存整合可信度的框架。 TrustMem relies on a Memory Transition Verifier to evaluate the transition process of memory updates in terms of coverage, preservation, and faithfulness. It further constructs preference pairs among candidate updates under the same memory state, enabling preference-guided reinforcement learning to directly optimize memory updating behaviors. Extensive experiments demonstrate that TrustMem improves both memory utility and reliability: it achieves state-of-the-art results across MemoryAgentBench, HaluMem, and the Mem-alpha validation set, improves HaluMem memory extraction by 12.14 F1 points, and reduces transition-level omission, corruption, and hallucination by 40.1\%, 79.1\%, and 50.0\%, respectively, compared with the strongest baseline for each error类型。