Google DeepMind 论文发现自进化智能体边学边忘，学得最快的方法破坏旧任务最多
New Google DeepMind paper finds that self-improving agents forget as they learn, and the method that learned new tasks fastest broke old ones the most.
Google DeepMind 新论文发现，自进化智能体在编辑系统提示词、记忆或技能文件来学习新任务时，会遗忘此前已掌握的任务。团队在话题、工作流或领域不断变化的任务流上测试了 ReasoningBank、GEPA 和 Meta-Harness，其中 GEPA 学得最多，平均每块提升 6.2 分，但在其他任务上损失也最大，为 4.3 分；它学会向用户追问缺失信息后，后续任务块改写了这条规则，15.5 分的增益降至 2.5 分。作者建议，若智能体会自行修改提示词或记忆，应把任务专属规则分开存放，并在每次更新后重测旧任务。
推荐理由
AI 评分 62
论文用任务流实验量化了自进化智能体学新忘旧的取舍，对自改提示词或记忆的实践有直接参考。
主题
正文
当前显示原文，中文全文译文尚未就绪。
New Google DeepMind paper finds that self-improving agents forget as they learn, and the method that learned new tasks fastest broke old ones the most.
Self-evolving agents improve without retraining by editing their system prompt, memory, or skill files. In real use tasks keep shifting, and each edit made for 1 task also applies to every task that follows.
The team ran ReasoningBank, GEPA, and Meta-Harness on task streams that changed topic, workflow, or domain. They tracked how much each method learned on the current task and lost on the others.
GEPA learned the most, 6.2 points per block on average, and also lost the most elsewhere, 4.3 points. After it learned to ask users for missing details, a later block rewrote that rule and its 15.5-point gain fell to 2.5.
If your agent edits its own prompt or memory, keep task-specific rules separate and retest old tasks after every update.