From Debate to Equilibrium: Belief-Driven Multi-Agent LLM Reasoning via Bayesian Nash Equilibrium

• URL: http://arxiv.org/abs/2506.08292v1
• Date: Mon, 09 Jun 2025 23:49:14 GMT
• Title: From Debate to Equilibrium: Belief-Driven Multi-Agent LLM Reasoning via Bayesian Nash Equilibrium
• Abstract summary: Multi-agent frameworks can boost the reasoning power of large language models (LLMs), but they typically incur heavy computational costs and lack convergence guarantees.<n>We recast multi-LLM coordination as an incomplete-information game and seek a Bayesian Nash equilibrium (BNE)<n>We introduce Efficient Coordination via Nash Equilibrium (ECON), a hierarchical reinforcement-learning paradigm that marries distributed reasoning with centralized final output.
• Score: 52.28048367430481
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multi-agent frameworks can substantially boost the reasoning power of large language models (LLMs), but they typically incur heavy computational costs and lack convergence guarantees. To overcome these challenges, we recast multi-LLM coordination as an incomplete-information game and seek a Bayesian Nash equilibrium (BNE), in which each agent optimally responds to its probabilistic beliefs about the strategies of others. We introduce Efficient Coordination via Nash Equilibrium (ECON), a hierarchical reinforcement-learning paradigm that marries distributed reasoning with centralized final output. Under ECON, each LLM independently selects responses that maximize its expected reward, conditioned on its beliefs about co-agents, without requiring costly inter-agent exchanges. We mathematically prove that ECON attains a markedly tighter regret bound than non-equilibrium multi-agent schemes. Empirically, ECON outperforms existing multi-LLM approaches by 11.2% on average across six benchmarks spanning complex reasoning and planning tasks. Further experiments demonstrate ECON's ability to flexibly incorporate additional models, confirming its scalability and paving the way toward larger, more powerful multi-LLM ensembles. The code is publicly available at: https://github.com/tmlr-group/ECON.

Related papers

• Attributing Emergence in Million-Agent Systems [68.53670424791751]
  Large language models (LLMs) can simulate human-like reasoning and decision-making in individual agents.<n>Such studies require attributing macro emergence to individual agents.<n>We adapt Aumann--Shapley path-integral attribution to LLM-powered MAS at million-agent scale.
  arXiv  Detail & Related papers  (2026-05-12T01:49:41Z)
• Why Keep Your Doubts to Yourself? Trading Visual Uncertainties in Multi-Agent Bandit Systems [21.356119126402902]
  We introduce Agora, a framework that reframes coordination as a decentralized market for uncertainty.<n>A market-aware broker, extending Thompson Sampling, initiates collaboration and guides the system toward cost-efficient equilibria.<n>Results establish market-based coordination as a principled and scalable paradigm for building economically viable visual intelligence systems.
  arXiv  Detail & Related papers  (2026-01-26T17:58:53Z)
• Mechanism-Based Intelligence (MBI): Differentiable Incentives for Rational Coordination and Guaranteed Alignment in Multi-Agent Systems [0.0]
  I introduce Mechanism-Based Intelligence (MBI), a paradigm that reconceptualizes intelligence as emergent from the coordination of multiple "brains", rather than a single one.<n>It provides a provably efficient, auditable and generalizable approach to coordinated, trustworthy and scalable multi-agent intelligence grounded in economic principles.
  arXiv  Detail & Related papers  (2025-12-22T22:22:13Z)
• Multimodal Reinforcement Learning with Agentic Verifier for AI Agents [131.46008226323423]
  Argos is a principled multimodal reward agent to train reasoning models for agentic tasks.<n>By leveraging our agentic verifier across both SFT data and RL training, our model achieves state-of-the-art results.
  arXiv  Detail & Related papers  (2025-12-03T04:42:47Z)
• Maestro: Learning to Collaborate via Conditional Listwise Policy Optimization for Multi-Agent LLMs [23.590034731179824]
  We present Through Role Orchestration (Maestro), a principled paradigm for collaboration that structurally decouples cognitive modes.<n>Maestro uses a collective of parallel Execution Agents for diverse exploration and a specialized Central Agent for convergent, evaluative synthesis.<n>Experiments on mathematical reasoning and general problem-solving benchmarks demonstrate that Maestro, coupled with CLPO, consistently outperforms existing state-of-the-art multi-agent approaches.
  arXiv  Detail & Related papers  (2025-11-08T21:01:27Z)
• Controlling Performance and Budget of a Centralized Multi-agent LLM System with Reinforcement Learning [53.57360296655208]
  Large language models (LLMs) exhibit complementary strengths across domains and come with varying inference costs.<n>Existing approaches rely on decentralized frameworks, which invoke multiple LLMs for every input and thus lead to substantial and uncontrolled inference costs.<n>We introduce a centralized multi-LLM framework, where a controller LLM selectively coordinates a pool of expert models in a cost-efficient and cost-controllable manner.
  arXiv  Detail & Related papers  (2025-11-04T17:35:17Z)
• Everyone Contributes! Incentivizing Strategic Cooperation in Multi-LLM Systems via Sequential Public Goods Games [4.3891974840097925]
  Multi-Agent Cooperation Sequential Public Goods Game (MAC-SPGG)<n>We introduce a novel, game-theoretically grounded reinforcement learning framework, to systematically incentivize cooperation in multi-LLM ensembles.<n>Our results highlight the power of structured, incentive-aligned MAC-SPGG cooperation for scalable and robust multi-agent language generation.
  arXiv  Detail & Related papers  (2025-08-04T05:36:07Z)
• Reasoning Like an Economist: Post-Training on Economic Problems Induces Strategic Generalization in LLMs [25.067282214293904]
  This paper explores whether post-training techniques, specifically Supervised Fine-Tuning (SFT) and Reinforcement Learning with Verifiable Rewards (RLVR), can effectively $textitgeneralize$ to multi-agent scenarios.<n>We use economic reasoning as a testbed, leveraging its strong foundations in mathematics and game theory.<n> Comprehensive evaluation on economic reasoning benchmarks and multi-agent games reveals clear improvements in structured reasoning and economic rationality.
  arXiv  Detail & Related papers  (2025-05-31T14:22:40Z)
• Why Ask One When You Can Ask $k$? Two-Stage Learning-to-Defer to the Top-$k$ Experts [3.6787328174619254]
  We introduce the first framework for Top-$k$ Learning-to-Defer, enabling systems to defer each query to the $k$ most cost-effective experts.<n>We propose Top-$k(x)$ Learning-to-Defer, an adaptive extension that learns the optimal number of experts per query based on input complexity, expert quality, and consultation cost.
  arXiv  Detail & Related papers  (2025-04-17T14:50:40Z)
• Vairiational Stochastic Games [1.6703448188585752]
  We propose a novel variational inference framework tailored to decentralized multi-agent systems.<n>Our framework addresses the challenges posed by non-stationarity and unaligned agent objectives.<n>We demonstrate theoretical convergence guarantees for the proposed decentralized algorithms.
  arXiv  Detail & Related papers  (2025-03-08T03:21:23Z)
• Symbolic Mixture-of-Experts: Adaptive Skill-based Routing for Heterogeneous Reasoning [76.10639521319382]
  We propose Symbolic-MoE, a symbolic, text-based, and gradient-free Mixture-of-Experts framework.<n>We show Symbolic-MoE beats strong LLMs like GPT4o-mini, as well as multi-agent approaches, with an absolute avg. gain of 8.15% over the best multi-agent baseline.
  arXiv  Detail & Related papers  (2025-03-07T18:03:13Z)
• Principal-Agent Reinforcement Learning: Orchestrating AI Agents with Contracts [20.8288955218712]
  We propose a framework where a principal guides an agent in a Markov Decision Process (MDP) using a series of contracts. We present and analyze a meta-algorithm that iteratively optimize the policies of the principal and agent. We then scale our algorithm with deep Q-learning and analyze its convergence in the presence of approximation error.
  arXiv  Detail & Related papers  (2024-07-25T14:28:58Z)
• Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents [101.17919953243107]
  GovSim is a generative simulation platform designed to study strategic interactions and cooperative decision-making in large language models (LLMs)<n>We find that all but the most powerful LLM agents fail to achieve a sustainable equilibrium in GovSim, with the highest survival rate below 54%.<n>We show that agents that leverage "Universalization"-based reasoning, a theory of moral thinking, are able to achieve significantly better sustainability.
  arXiv  Detail & Related papers  (2024-04-25T15:59:16Z)
• ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs [61.07130026622437]
  Large Language Models (LLMs) still struggle with natural language reasoning tasks. Motivated by the society of minds, we propose ReConcile. A multi-model multi-agent framework designed as a round table conference among diverse LLM agents.
  arXiv  Detail & Related papers  (2023-09-22T17:12:45Z)
• Maximum Entropy Heterogeneous-Agent Reinforcement Learning [45.377385280485065]
  Multi-agent reinforcement learning (MARL) has been shown effective for cooperative games in recent years.<n>We propose a unified framework for learning policies to resolve issues related to sample complexity, training instability, and the risk of converging to a suboptimal Nash Equilibrium.<n>Based on the MaxEnt framework, we propose Heterogeneous-Agent Soft Actor-Critic (HASAC) algorithm.<n>We evaluate HASAC on six benchmarks: Bi-DexHands, Multi-Agent MuJoCo, StarCraft Challenge, Google Research Football, Multi-Agent Particle Environment, and Light Aircraft Game.
  arXiv  Detail & Related papers  (2023-06-19T06:22:02Z)
• Inducing Stackelberg Equilibrium through Spatio-Temporal Sequential Decision-Making in Multi-Agent Reinforcement Learning [17.101534531286298]
  We construct a Nash-level policy model based on a conditional hypernetwork shared by all agents. This approach allows for asymmetric training with symmetric execution, with each agent responding optimally conditioned on the decisions made by superior agents. Experiments demonstrate that our method effectively converges to the SE policies in repeated matrix game scenarios.
  arXiv  Detail & Related papers  (2023-04-20T14:47:54Z)
• Efficient Model-based Multi-agent Reinforcement Learning via Optimistic Equilibrium Computation [93.52573037053449]
  H-MARL (Hallucinated Multi-Agent Reinforcement Learning) learns successful equilibrium policies after a few interactions with the environment. We demonstrate our approach experimentally on an autonomous driving simulation benchmark.
  arXiv  Detail & Related papers  (2022-03-14T17:24:03Z)
• Finding General Equilibria in Many-Agent Economic Simulations Using Deep Reinforcement Learning [72.23843557783533]
  We show that deep reinforcement learning can discover stable solutions that are epsilon-Nash equilibria for a meta-game over agent types. Our approach is more flexible and does not need unrealistic assumptions, e.g., market clearing. We demonstrate our approach in real-business-cycle models, a representative family of DGE models, with 100 worker-consumers, 10 firms, and a government who taxes and redistributes.
  arXiv  Detail & Related papers  (2022-01-03T17:00:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.