SOM: Structured Opponent Modeling for LLM-based Agents via Structural Causal Model

2026-05-08

AI Summary

This work addresses the limitations of existing large language model (LLM)-based multi-agent approaches, which struggle to accurately and flexibly predict opponent behavior due to tightly coupled modeling and prediction processes, reliance on implicit reasoning, and limited adaptability. To overcome these challenges, the paper introduces Structured Opponent Modeling (SOM), a novel framework that incorporates structural causal models (SCMs) to explicitly construct causal graphs of opponents’ behaviors and guide LLMs through structured, causal-path-based reasoning. By decoupling modeling from prediction, SOM enables interpretable, transferable, and dynamically adaptive inference of opponent strategies. Empirical evaluations across multiple multi-agent benchmarks demonstrate that SOM significantly outperforms state-of-the-art LLM baselines, achieving substantial improvements in both prediction accuracy and strategic stability.

Abstract

Accurately predicting opponents' behavior from interactions is a fundamental capability for large language model (LLM)-based agents in multi-agent and game-theoretic environments. Existing approaches often entangle opponent modeling with prediction, relying on implicit contextual reasoning and limiting adaptability in dynamic interactions. To this end, we propose Structured Opponent Modeling (SOM), a two-stage opponent modeling framework that distinctly separates opponent model construction and opponent prediction. At the construction stage, SOM employs a Structural Causal Model (SCM), a graph-based formalism for representing dependencies among variables, to capture directed links between opponents' observations and actions, yielding an explicit and structured opponent representation. At the prediction stage, the LLM performs structured reasoning along clear pathways derived from the SCM, improving both prediction accuracy and stability. Extensive experiments on diverse multi-agent benchmarks demonstrate that SOM consistently outperforms state-of-the-art LLM-based reasoning baselines, enabling more accurate and adaptable strategic decision-making in complex and dynamic multi-agent interactions.

Problem

• opponent modeling
• large language models
• multi-agent systems
• behavior prediction
• dynamic interactions

Innovation

• Structured Opponent Modeling
• Structural Causal Model
• LLM-based agents
• multi-agent reasoning
• causal representation