# When Harnesses Lose the Signal: Causal Evaluation of Recovery in LLM Agents

> 全文为英语

**Published:** 2026年10月2日
**Author:** Shuyao Xiao, Shengling Wang, Xuan Chen, Ke Chao, Ming Cui, Feifei Qian, Chaoyang Mei, Fanlin Meng, Ziming Yu, Junxi Yin
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/zh/articles/when-harnesses-lose-the-signal-causal-evaluation-of-recovery-in-llm-agen

Large language model agents rely on external harnesses to pass information between the model and its environment and to recover from execution errors. Yet recovery is usually judged only by average task success. This hides an important tension. The same operation can rescue a…

> 内容来自原始来源： **arXiv — cs.AI**
> [在 arXiv — cs.AI 阅读全文](https://arxiv.org/abs/2610.00372)

**标签：** #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*