重复多方 LLM 谈判研究揭示「信任—协议鸿沟」：识别对手不等于达成协议
Knowing Whom to Trust is not Enough: The Trust--Agreement Gap in Repeated Multi-Party LLM Negotiation
AI 导读
一项 arXiv 研究构建了重复多方多议题 LLM 谈判框架，其中一名参与者暗中阻止达成协议，智能体在每轮谈判后维护成对信任评估。结果显示存在「信任—协议鸿沟」：信任感知智能体几轮内就能识别出对手，但准确的信任估计并不必然带来更好的协议。只有当智能体被明确引导在决策中使用信任时，协议率才最多提升 23 个百分点。
正文 · 原文
Abstract:Large language model (LLM) agents increasingly participate in negotiations on behalf of individuals and organizations, yet they remain vulnerable to manipulation, particularly in repeated multi-party interactions. Classical multi-agent systems address this challenge through explicit trust and reputation models, whereas existing work on LLM negotiation largely treats trust as an emergent phenomenon. We investigate whether LLM agents can form useful trust estimates from prior interactions and leverage them to improve future negotiation outcomes. We introduce a repeated, multi-party, multi-issue negotiation framework in which recurring agents participate in a sequence of negotiation games while one participant covertly attempts to prevent agreement. After each game, agents assess their counterparts and maintain pairwise trust estimates that carry over to subsequent negotiations. We compare trust-aware and trust-agnostic agents and analyze how explicit trust influences negotiation outcomes. Our results reveal a \emph{trust--agreement gap}: trust-aware agents can identify the adversary within a few games, but accurate trust estimates do not consistently translate into better agreements. Trust increases agreement rates by up to 23 percentage points, but only when agents are explicitly guided to use it in their decisions. Explicit trust can therefore improve repeated LLM negotiations, but its value depends critically on how trust is assessed and incorporated into agents' reasoning.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2610.31971 [cs.AI] |
| (or arXiv:2610.31971v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.31971 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Amit Sarker [view email]
[v1]
Thu, 22 Oct 2026 22:24:10 UTC (698 KB)
来源：arXiv cs.AI · arxiv.org