Can Agent Memory Systems Track Evolving State?

LLM agent 记忆系统在长交互中面临状态演化问题：约束变化、决策被覆盖，但现有基准测的是静态回忆

新基准：构造含状态演化的多轮交互场景，测量 agent 记忆系统能否反映最新状态而非历史累加

揭示主流记忆系统在状态跟踪上的系统性缺陷；提出评测记忆「新鲜度」的新维度

全双工 agent 中对话状态实时变化，需要记忆系统持续追踪而非静态积累

As LLM-based agents are deployed for longer and higher-stakes tasks, their memory systems continue to have crucial gaps. While existing memory benchmarks focus largely on recall-shaped tasks, we argue an effective memory system must track the evolving state of the world; as facts, constraints, and decisions are revised over a long interaction, answers must reflect the current state and not a superseded one. We define this capability as state tracking and instantiate it in StateMemBench, a benchm