AWS 与 HSBC 提出 Double Ratchet：仅需 10 个标注样本即可自动进化 Agent 评分器
给 Agent 写评分器最烦，AWS 和 HSBC 这篇论文说 10 个标注样本就够自动进化评分器，做 Agent 自我改进的可以看看。
AWS 与 HSBC 在论文中提出 Double Ratchet 方法，让 Agent 自我改进所需的评分器与技能同步进化，无需人工编写完整评分标准。评分器由针对单一失败类型的规则化检查构成，其版本只依据与 10 个标注样本的一致性来选择，不会参考 Agent 得分是否提升。在代码、SQL、报告三类任务上，该方法达到真实标签或人工评分标准效果的 88% 至 110%。移除防作弊机制后评分器虽对 97% 至 100% 的输出放行，但训练出的技能效果相当。
Self-improving agents have always needed a human to write the grader.
A new AWS + HSBC paper removes that step: the grader evolves alongside the agent from 10 labeled examples and trains it about as well as ground truth.
Double Ratchet evolves the grader alongside the agent's skills, building it from small, mostly rule-based checks for 1 failure type each. A grader version is chosen for agreeing with 10 labeled examples, never for raising the agent's score.
It kept 88% to 110% of the gain that ground truth or a rubric gave the same skill loop, across code, SQL, and reports. With the safeguards removed, the grader passed 97% to 100% of outputs yet trained skills as well.
