用真理欺骗：通过生成蒙太奇进行开放式通道多智能体合谋以操纵信念
Lying with Truths: Open-Channel Multi-Agent Collusion for Belief Manipulation via Generative Montage
AI总结：
本文研究了通过公开通道分发真实证据片段，利用多智能体合谋操纵信念的新威胁，提出了生成蒙太奇框架，展示了在14种LLM家族中74.4%的攻击成功率，并揭示了更强的推理能力反而增加了易受攻击的风险。
摘要：
随着大型语言模型（LLMs）向自主代理合成实时信息转变，其推理能力引入了意想不到的攻击面。本文介绍了一种新的威胁，即合谋代理通过仅使用真实证据片段在公开通道中引导受害者信念，而无需依赖隐蔽通信、后门或伪造文件。通过利用LLMs的过度思考倾向，我们正式化了首次认知合谋攻击，并提出生成蒙太奇：一个由写作者-编辑-导演框架构成的框架，通过对抗性辩论和协调发布证据片段来构建欺骗性叙述，使受害者内化并传播伪造结论。为研究此风险，我们开发了CoPHEME数据集，该数据集源自真实世界谣言事件，并在多种LLM家族中模拟攻击。我们的结果表明，14种LLM家族普遍存在漏洞：攻击成功率达到74.4%（专有模型）和70.6%（开放式权重模型）。反直觉的是，更强的推理能力增加了易受攻击性，推理专精模型的攻击成功率高于基础模型或提示。此外，这些虚假信念会传播到下游判断者，达到超过60%的欺骗率，突显了LLM代理在动态信息环境中交互的社会技术脆弱性。我们的实现和数据可在：https://github.com/CharlesJW222/Lying_with_Truth/tree/main。
英文摘要：
As large language models (LLMs) transition to autonomous agents synthesizing real-time information, their reasoning capabilities introduce an unexpected attack surface. This paper introduces a novel threat where colluding agents steer victim beliefs using only truthful evidence fragments distributed through public channels, without relying on covert communications, backdoors, or falsified documents. By exploiting LLMs' overthinking tendency, we formalize the first cognitive collusion attack and propose Generative Montage: a Writer-Editor-Director framework that constructs deceptive narratives through adversarial debate and coordinated posting of evidence fragments, causing victims to internalize and propagate fabricated conclusions. To study this risk, we develop CoPHEME, a dataset derived from real-world rumor events, and simulate attacks across diverse LLM families. Our results show pervasive vulnerability across 14 LLM families: attack success rates reach 74.4% for proprietary models and 70.6% for open-weights models. Counterintuitively, stronger reasoning capabilities increase susceptibility, with reasoning-specialized models showing higher attack success than base models or prompts. Furthermore, these false beliefs then cascade to downstream judges, achieving over 60% deception rates, highlighting a socio-technical vulnerability in how LLM-based agents interact with dynamic information environments. Our implementation and data are available at: https://github.com/CharlesJW222/Lying_with_Truth/tree/main.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
