自改进智能体的 harness 编辑准入规则
Self-improving agents should accept a harness edit only after it beats the current version on fresh, independent tasks.
自改进智能体应仅在 harness 编辑在新独立任务上击败当前版本且达到设定幅度时才被采纳，该准入规则将 harness 自改进方法映射到进化算法，涵盖保留最优搜索、档案、Pareto 选择与 bandit 引导预算。案例中 GPT-5.4-nano 搭配进化后的 harness 与技能在 SpreadsheetBench 得 77.58%，GPT-5.5 仅做技能调优得 76.20%，但成本未对齐。若智能体可改自身配置，应在全新任务上测试每次改动，并将评估器、权限与紧急停止置于其触及范围之外。
正文
当前显示原文，中文全文译文尚未就绪。
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.