诚实谎言：理解反思型代理中的记忆虚构

摘要

反思式智能体依赖自我生成的反思作为记忆，隐含假设智能体能够准确诊断自身失败。我们发现这一假设可能系统性失效：在ALFWorld和HumanEval中，智能体会存储对任务的自信但错误的解读，并在多次试验中持续基于这些解读行动，即便环境每次都重置为正确任务。我们将这种失败模式称为**记忆虚构**，并引入**反思重复率（RRR）**，一种基于日志的度量指标，用于检测对错误反思内容的重复依赖。通过RRR，我们在ALFWorld中识别出16个冻结环境，其中121条反思中无一提及正确目标对象，在HumanEval中也发现4个类似案例。我们的缓解方案将开放式自我诊断替换为程序化提取轨迹级别的失败信号，使正确对象提及率从0%提升至86%，RRR从0.64降至0.10，并成功解决了ALFWorld中16个冻结环境中的3个，这表明反思记忆可能强化而非纠正错误信念。

English

Reflexion-style agents rely on self-generated reflections as memory, implicitly assuming that agents can accurately diagnose their own failures. We show that this assumption can fail systematically: across ALFWorld and HumanEval, agents store confident but incorrect interpretations of the task and continue acting on them across trials, even though the environment resets to the correct task each time. We call this failure mode memory confabulation and introduce the Reflection Repetition Rate (RRR), a log-based metric that detects repeated reliance on incorrect reflective content. Using RRR, we identify 16 frozen environments in ALFWorld, where 0 of 121 reflections mention the correct target object, and 4 analogous cases in HumanEval. Our mitigation replaces open-ended self-diagnosis with programmatic extraction of trajectory-level failure signals, increasing correct object mention from 0% to 86%, reducing RRR from 0.64 to 0.10, and solving 3 of 16 frozen ALFWorld environments, suggesting that reflective memory can reinforce false beliefs rather than correct them.