AI 热点

Google 与宾州州立大学论文提出 SIM,为智能体框架改动记录理由

Rohan Paul更新于 11月18日 05:02

New Google and Penn State paper finds that when one AI rewrites the code that gives another AI its instructions and tools, keeping notes on why each change was

AI 导读

Google 与宾州州立大学的新论文提出 Self-Improvement Memory(SIM),在让一个 AI 改写另一个 AI 的指令与工具代码时,记录每次改动针对的问题、推理过程和实测结果。把 Claude Opus 5.5 的 harness 迁移到 Gemini 3.8 Flash 的 workspace 任务时,SIM 将分数从 68.41 提升到 84.75,而无日志时为 73.42;在 10 次重复编码运行中,SIM 有 9 次领先,平均高出 2.36 分。

推荐理由

AI 评分 57

论文给出跨模型迁移时保留改动理由的做法,并用一组具体分数对比说明日志带来的差异。

主题

标签

正文

New Google and Penn State paper finds that when one AI rewrites the code that gives another AI its instructions and tools, keeping notes on why each change was made helps the gains carry over to a new model.

LLMs can now rewrite an agent's harness, the code that runs prompts and tools. After a model or benchmark switch, it can score below the original, and its code hides why edits were made.

Self-Improvement Memory, or SIM, logs each edit with the problem it targeted, its reasoning and its measured result. The optimizer reads this log on the new setup to decide which old edits to keep or revise.

When a Claude Opus 5.5 harness moved to Gemini 3.8 Flash workspace tasks, SIM lifted it from 68.41 to 84.75, versus 73.42 without the log. Across 10 repeated coding runs, SIM finished ahead 9 times, by 2.36 points on average.

Store why each harness edit was made and how it scored, and carry that log to new models.

来源

Rohan Paul