# Proxy Confidence: Auditing Black-Box LLM Agents with a Surrogate's Log-Probabilities

> Толық мәтін ағылшын тілінде

**Published:** 2026 ж. 6 қазан
**Author:** Yikai Zhao, Saurabh Pandey, Pradeep Kumar Misra
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/kk/articles/proxy-confidence-auditing-black-box-llm-agents-with-a-surrogate-s-log-pr

A deployed LLM agent emits tool calls, queries, and code that can be silently wrong -- by the time the error surfaces, the action has run. Frontier chat APIs hide the model's token probabilities; the agent's stated confidence barely beats chance on the mistakes that matter; and…

> Материал бастапқы дереккөзден алынған: **arXiv — cs.AI**
> [Толық мәтінді arXiv — cs.AI сайтында оқу](https://arxiv.org/abs/2610.03894)

**Тегтер:** #arXiv

---

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