大语言模型智能体助力 One Health 风险协商与共识构建研究
Towards using Large Language Models for Risk Negotiation and Consensus-building in One Health contexts
一项 arXiv 研究提出将大语言模型智能体融入人工监督谈判流程的多步骤管线，用于 One Health 框架下的风险评估、估值与协商。智能体基于 RAG 从利益相关方材料生成，结合领域计算模型与网络来源，模拟谈判结果用 Nash 讨价还价解与 Pareto 最优指标刻画。在禽肉弯曲杆菌食品安全标准与橡树蛾管理两个真实场景的专家工作坊中，参与者均在限定时间内达成共识。
Authors:Alexandra Fetsch, Iurii Savvateev, Racem Ben Romdhane, Martin Wiedmann, Artemiy Dimov, Maciej Durkalec, Josef Teichmann, Jakob Zinsstag, Konstantinos Koutsoumanis, Andreja Rajkovic, Jason Mann, Mauro Tonolla, Monika Ehling-Schulz, Matthias Filter, Sophia Johler
Abstract:Risk negotiation under the One Health framework requires balancing competing stakeholder interests across human, animal, and environmental health domains - a process constrained by information complexity and time pressure. This study presents and evaluates a multi-step pipeline that integrates large language model-based agents into a human-supervised negotiation process, specifically targeting risk assessment, valuation, and negotiation steps. Agent descriptions were generated from stakeholder-provided materials using retrieval-augmented generation methods. During preliminary multi-agent discussions, agents accessed domain-specific computational models and web-based sources before human participants refined proposed issues and options. A simulated negotiation game then produced distributions of candidate deals, characterised using Nash bargaining solution criteria and Pareto optimality metrics. The developed approach was evaluated in a workshop involving independent domain experts across two real-world scenarios: food safety criteria for Campylobacter in broiler meat and oak processionary moth management. In this feasibility evaluation, participants reached consensus with the allotted time after discussing and modifying the proposed deals. The framework offers an adaptable blueprint for AI-assisted risk negotiation; the specific contribution of LLM assistance, including gains in efficiency and decision quality, requires comparative evaluation.
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2509.09906 [cs.MA] |
| (or arXiv:2509.09906v2 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2509.09906 arXiv-issued DOI via DataCite |
Submission history
From: Alexandra Fetsch [view email]
[v1]
Fri, 12 Sep 2025 00:25:20 UTC (16,739 KB)
[v2]
Tue, 10 Nov 2026 08:49:19 UTC (7,679 KB)
来源：arXiv cs.AI · arxiv.org