做AI智能体开发的团队会立刻警觉——你精心设计的记忆系统可能在悄悄退化。这篇论文用实验数据戳破了「自动总结记忆」的幻觉，建议所有用LLM做长期任务的开发者点开看看，别让记忆成为瓶颈。
伊利诺伊大学和清华大学等实验室的研究发现，LLM智能体在持续重写自身记忆时，记忆会变得不可靠。许多智能体系统通过让LLM将混乱经验压缩成整洁教训来存储过往工作，但论文表明反复重写会逐渐损害记忆。原始经验（实际尝试和解决方案）往往比精炼后的教训更有用。在网页购物、模拟世界、应用使用和ARC-AGI谜题等任务中测试，GPT-5.4在无记忆时解决100%的ARC-AGI任务，但用正确解决方案构建记忆后，流式更新导致准确率降至约54%。失败源于错误分组、过度泛化和过拟合，记忆丢失细节、混淆任务类型或学习到仅适用于狭窄示例的规则。论文建议智能体记忆不应自动将每次经验重写为摘要，保留原始证据并仅偶尔制作摘要效果更好。
This Illinois+ Tsinghua University and other labs …
This Illinois+ Tsinghua University and other labs study finds that LLM agents still have unreliable memory and that it can get worse when they keep rewriting their own memories.
LLM agents can learn from experience, but their rewritten memories often become unreliable.
The problem is that many agent systems store past work by asking an LLM to compress messy experience into neat written lessons.
That sounds useful because the agent should remember what worked before, but the paper finds that repeated rewriting slowly damages the memory.
The core idea is that raw episodes, meaning the actual past attempts and solutions, often stay more useful than the polished lessons made from them.
The authors tested this across tasks like web shopping, simulated worlds, app use, and ARC-style puzzle problems where they could control the correct solutions.
The sharpest result is that GPT-5.4 solved 100% of a small ARC-AGI set with no memory, but after memory was built from correct solutions, streaming updates dropped it to about 54%.
The failures came from bad grouping, overbroad lessons, and overfitting, so the memory forgot details, mixed up task types, or learned rules that only worked on narrow examples.
The big deal is that agent memory should not automatically rewrite every experience into a summary, because keeping raw evidence and only sometimes making summaries worked better.
The paper is really proposing that agent memory should treat raw past episodes as important evidence, not as disposable notes to summarize away.
----
arxiv. org/abs/2605.12978
Title: "Useful Memories Become Faulty When Continuously Updated by LLMs"