---
id: 20261111-T0-09
title: "LLM智能体容易被骗？新框架用心智理论识别并抵抗欺骗"
title_en: "AVCORP Framework Helps LLM Agents Detect and Resist Deception"
url: https://ai.daily.yangsir.net/daily/20261111-T0-09
issue_date: 2026-11-11
publish_date: 2026-11-10T05:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2611.10176
---

# LLM智能体容易被骗？新框架用心智理论识别并抵抗欺骗

这篇论文关注LLM智能体容易被用户或其他智能体欺骗的问题，提出AVCORP框架。它的思路是结合心智理论（ToM）推理去推断对方的信念和目标，再用通信理论对欺骗策略建模，从而识别并抵御欺骗。论文发布在arXiv（编号2611.10176）。实际影响是，多智能体协作、客服、自动化交易这类需要判断对方意图的场景，可以借助该框架降低被诱导或操纵的风险。

## English Version

**AVCORP Framework Helps LLM Agents Detect and Resist Deception**

LLM-based agents are vulnerable to deception from users and other agents. A new arXiv paper (2611.10176) introduces AVCORP, a framework that combines Theory of Mind reasoning, inferring others' beliefs and goals, with communication theory to model deceptive tactics, letting agents identify and resist deception. The work targets multi-agent collaboration, customer service and automated trading, where misreading intent can lead to manipulation.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2611.10176)

**详情页**：https://ai.daily.yangsir.net/daily/20261111-T0-09

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*