# Hacking or Just Failing? A More Capable LLM Monitor Better Separates Reward Hacking from Honest Failure

> 全文为英语

**Published:** 2026年11月3日
**Author:** Kaushik Kezendirane
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/zh/articles/hacking-or-just-failing-a-more-capable-llm-monitor-better-separates-rewa

arXiv:2611.00407v1 Announce Type: new Abstract: An agent's visible score can exceed its real progress because it gamed the score or because it simply made mistakes, so a useful reward-hacking monitor must tell the two apart. We built a sandboxed task in which an agent can game a…

> 内容来自原始来源： **arXiv — cs.AI**
> [在 arXiv — cs.AI 阅读全文](https://arxiv.org/abs/2611.00407)

**标签：** #Google DeepMind #arXiv #Meta AI

---

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