多智能体商议（MAD）如何影响人机协作决策
Understanding the Effects of Multi-Agent Deliberation for Human-AI Decision-Making
AI 导读
arXiv 论文通过 LLM 回答验证与事实核查两项用户实验，检验多智能体商议（MAD）对人类决策的帮助。结果显示 MAD 优于单轮 AI 辅助，其中多轮交互是主要驱动因素，第二个 AI 仅在单个 AI 有改进空间时才有增益。此外，魔鬼代言人角色会降低准确度与用户体验，而人类积极参与仅在一项任务中提高准确度但增加了挫败感。
正文 · 原文
Abstract:Multi-agent deliberation (MAD), where multiple LLMs engage in structured discussion, improves AI performance on many benchmarks, but whether it helps humans make better decisions remains untested. We conduct two user studies on LLM-response-verification and fact-checking. We first compare MAD against common AI decision-support methods that isolate its two defining dimensions --- multiple AIs and multi-round interaction. We find MAD outperformed single-round AI assistance on both tasks; multi-round interaction was the main driver of this benefit, while a second AI added value when the single AI had room to improve. Then, we study two design choices in MAD --- AI persona and human engagement level --- finding a devil's advocate persona to lower accuracy and user experience, while humans' active participation in MAD improved accuracy in one task at the cost of frustration. We conclude by discussing implications for when, why, and how multi-agent deliberation can support human decision-making.
| Subjects: | Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.14901 [cs.HC] |
| (or arXiv:2611.14901v1 [cs.HC] for this version) | |
| https://doi.org/10.48550/arXiv.2611.14901 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Harry Tian [view email]
[v1]
Wed, 11 Nov 2026 01:42:31 UTC (1,110 KB)
来源：arXiv cs.AI · arxiv.org