Inducing Reward-Free Judging Rubrics that Reduce Over-Crediting in Agent Evaluation

论文摘要：Evaluating language-model agents at scale increasingly relies on a second language model as an automatic judge, because the gold signal, an executable environment reward, is e...

为什么重要

该论文来自一手研究源并通过 3qk7 Research Radar 的新鲜度、来源与研究价值筛选；后续代码、模型权重、复现和相关产品事件可继续关联到该记录。

检测到的变化

• 研究论文进入 3qk7 高价值 Research Radar

一手证据

• arXiv cs.AI · 2026-08-17T22:05:17Z