# 以辩欺骗：通过博弈论多智能体辩论毒化LLM的事实核查

> 全文为英语

**Published:** 2026年11月9日
**Author:** Liting Huang, Huizhi Liang, Yi Lei, Xingjun Ma, Varun Ojha
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/zh/articles/debate-to-deceive-poisoning-llm-s-fact-checking-via-game-theoretic-multi

arXiv:2611.08756v1 公告类型：new 摘要：针对大语言模型的知识攻击，如操纵事实核查判断，对AI可靠性构成重大风险。现有方法——对抗性后缀/模板注入和单智能体自我反思——…

arXiv:2611.08756v1 Announce Type: new Abstract: Knowledge attacks on Large Language Models, such as manipulating fact-checking judgments, pose significant risks to AI reliability. Existing approaches -- adversarial suffix/template injection and single-agent self-reflection -- lack robustness across diverse defenses and are prone to confirmation bias and costly trial-and-error, respectively. We propose Debate-to-Deceive (D2D), a multi-agent debate framework grounded in human debate structure and rhetorical style, enabling agents to generate adversarial passages that are semantically coherent, rhetorically effective, and resistant to early confirmation bias. D2D formalizes the adversarial process as an evolutionary game in which agents adaptively evolve their rhetorical strategies (Logos, Pathos, Ethos) through competition. To manage the intractably large LLM output space, we decouple agent behavior into a strategic layer, which selects a rhetorical action from a compact strategy space via evolutionary-inspired learning dynamics, and a generation layer, where the LLM produces a concrete passage conditioned on that strategy and the debate context. On four fact-checking benchmarks (FEVER, HealthVer, CLIMATE-FEVER, and SciFact), D2D substantially outperforms existing attacks and evades most defenses.

---

*内容来自原始来源： [arXiv — cs.AI](https://arxiv.org/abs/2611.08756). 查看原文.*

**标签：** #arXiv

---

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