Preference Estimation via Opponent Modeling in Multi-Agent Negotiation

ACL ARR 2026 January Submission6313 Authors

Keywords: Automated Negotiation, Multi-Agent Systems, Opponent Modeling, Bayesian Learning, LLMs

Abstract: Automated negotiation is essential for facilitating decision-making in societies with diverse stakeholders. While opponent modeling is critical, particularly in multi-party, multi-issue settings, conventional numerical-only methods fail to incorporate the qualitative context embedded in natural language, leading to unstable predictions. Although Large Language Models (LLMs) enable context-aware reasoning, they often struggle to maintain inferential consistency during prolonged interactions. To bridge this gap, we propose a novel preference estimation method integrating natural language information into a structured Bayesian opponent modeling framework. Our approach leverages LLMs to extract qualitative cues from utterances and converts them into probabilistic formats for dynamic belief tracking. Experimental results using a multi-party benchmark demonstrate that our framework achieves high agreement rates among all participants and superior preference estimation accuracy. These findings suggest that the integration of structured probabilistic reasoning and natural language understanding can facilitate robust consensus-building in ambiguous social interactions.

Paper Type: Short

Research Area: AI/LLM Agents

Research Area Keywords: Autonomous agents, LLM agents, multi-agent systems, agent coordination and negotiation

Languages Studied: English

Submission Number: 6313