RuHarness 同步演化评估标准与智能体测试框架
RuHarness 提出同步演化评估标准与智能体测试框架，在留出任务上使用 Qwen3.5-35B-A3B 时比最强框架演化基线高 4.5 分，使用 GLM-4.7-Flash 时高 5.3 分。该方法从轨迹中寻找行为标准、重新评分，并让框架修改聚焦尚未解决的弱点；不再触发的标准会休眠，弱点重现时再启用。论文链接为 arxiv.org/abs/2610.36826。
Self-improving harnesses are one of the most discussed ideas in AI right now.
(bookmark it)
This work proposes a new harness, RuHarness, that beats the strongest harness-evolution baseline by 4.5 points on held-out tasks with Qwen3.5-35B-A3B, and by 5.3 points with GLM-4.7-Flash.
The work targets a gap in current methods. They evolve the harness but keep the criteria that guide each edit fixed, so the criteria stop matching the agent's weaknesses after a few rounds.
RuHarness evolves the rubric along with the harness. Each round, it finds behavioral criteria in the trajectories, re-scores them, and directs harness edits at the most relevant unresolved weakness. Criteria that stop firing go dormant and return if the weakness reappears.
If your harness evolution stops improving after a few rounds, update the critic too.
arxiv.org/abs/2610.36826
Chat with Paper: academy.dair.ai/papers/when-…
来源：dair_ai · x.com