论文提出自进化智能体应仅在独立新任务上验证通过后才采纳 harness 编辑
Self-evolving Agent Harness 论文提出自进化智能体应仅在独立新任务上验证通过后才采纳 harness 编辑
一篇论文提出，自进化智能体只有在新的独立任务上以一定幅度胜过当前版本时，才应采纳对 harness 的编辑，并将 harness 自改进方法映射到进化算法上。论文按编辑所改内容分类，涵盖提示词、记忆、工具与运行时代码，并警告被采纳的编辑会带入后续运行，可能过拟合、抹掉旧技能或扩散给其他用户。
Self-improving agents should accept a harness edit only after it beats the current version on fresh, independent tasks.
It maps harness self-improvement methods onto evolutionary algorithms, from simple keep-the-best search to archives, Pareto selection, and bandit-guided budgets.
Self-improving agents now rewrite their harness: the prompts, memory, tools, and code around a fixed model. Accepted edits carry into later runs, so a bad edit can overfit, erase old skills, or spread to other users.
They sort these edits by what they change, from prompts to runtime code. Their admission rule: keep an edit only if it beats the current harness on new, independent tasks by a set margin.
In a case they compile from prior work, GPT-5.4-nano with an evolved harness and skills scored 77.58% on SpreadsheetBench. GPT-5.5 with skill tuning alone scored 76.20%, though costs were not matched.
If your agent edits its own setup, test each change on fresh tasks, and keep evaluators, permissions, and emergency stops out of its reach.
来源：Rohan Paul · x.com