CoMMa: Contribution-Aware Medical Multi-Agents From A Game-Theoretic Perspective

TL;DR

CoMMa introduces a decentralized, contribution-aware multi-agent framework for oncology decision support, utilizing game theory and deterministic embeddings to improve interpretability, stability, and accuracy in complex medical tasks.

Contribution

It presents a novel decentralized multi-agent system that uses game-theoretic objectives and deterministic embeddings for contribution-aware decision-making in medical oncology.

Findings

Achieves higher accuracy than baseline models.

Provides explicit evidence attribution and interpretability.

Demonstrates stability across diverse oncology datasets.

Abstract

Recent multi-agent frameworks have broadened the ability to tackle oncology decision support tasks that require reasoning over dynamic, heterogeneous patient data. We propose Contribution-Aware Medical Multi-Agents (CoMMa), a decentralized LLM-agent framework in which specialists operate on partitioned evidence and coordinate through a game-theoretic objective for robust decision-making. In contrast to most agent architectures relying on stochastic narrative-based reasoning, CoMMa utilizes deterministic embedding projections to approximate contribution-aware credit assignment. This yields explicit evidence attribution by estimating each agent's marginal utility, producing interpretable and mathematically grounded decision pathways with improved stability. Evaluated on diverse oncology benchmarks, including a real-world multidisciplinary tumor board dataset, CoMMa achieves higher accuracy and more stable performance than data-centralized and role-based multi-agents baselines.