LinguaGame: A Linguistically Grounded Game-Theoretic Paradigm for Multi-Agent Dialogue Generation

2026-01-08

arXiv.org

AI Summary

This work addresses the inefficiency of natural language interaction in multi-agent systems by proposing the first linguistically grounded, task-agnostic game-theoretic dialogue framework, which models conversation as an intentional and strategic signaling game. Integrating linguistic reasoning with signaling game theory, the authors design a training-free, inference-time equilibrium approximation algorithm that dynamically optimizes agents’ dialogue strategies. Experimental evaluations in simulated courtroom and debate scenarios demonstrate that the proposed approach significantly enhances communicative efficiency and receives strong endorsement from human expert assessments.

Technology Category

• Multiagent Systems: Agent Communication
• Natural Language Processing: Conversational AI/Dialog Systems
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Large Language Models (LLMs) have enabled Multi-Agent Systems (MASs) where agents interact through natural language to solve complex tasks or simulate multi-party dialogues. Recent work on LLM-based MASs has mainly focused on architecture design, such as role assignment and workflow orchestration. In contrast, this paper targets the interaction process itself, aiming to improve agents'communication efficiency by helping them convey their intended meaning more effectively through language. To this end, we propose LinguaGame, a linguistically-grounded game-theoretic paradigm for multi-agent dialogue generation. Our approach models dialogue as a signalling game over communicative intents and strategies, solved with a training-free equilibrium approximation algorithm for inference-time decision adjustment. Unlike prior game-theoretic MASs, whose game designs are often tightly coupled with task-specific objectives, our framework relies on linguistically informed reasoning with minimal task-specific coupling. Specifically, it treats dialogue as intentional and strategic communication, requiring agents to infer what others aim to achieve (intents) and how they pursue those goals (strategies). We evaluate our framework in simulated courtroom proceedings and debates, with human expert assessments showing significant gains in communication efficiency.

Problem

• multi-agent dialogue
• communication efficiency
• linguistic grounding
• intentional communication
• strategic interaction

Innovation

• signalling game
• multi-agent dialogue
• linguistically grounded
• communication efficiency
• equilibrium approximation