Work

MAPRO: Recasting Multi-Agent Prompt Optimization as Maximum a Posteriori Inference

Year: 2026

Type: conference-paper

Abstract: Large language models (LLMs) have demonstrated remarkable capabilities across diverse tasks, and LLM-based agents further extend these abilities to various practical workflows.While recent progress shows that multi-agent systems (MAS) can outperform single agents by coordinating specialized roles, designing effective MAS remains difficult due to prompt sensitivity and the compounded instability MAS creates.To cope with the challenge, recent efforts in automated prompt design have reduced manual effort.However, multi-agent prompt optimization remains largely unexplored.Challenges like exponentially expanding search space and ambiguous credit assignment together make systematic design intractable without principled methods.Therefore, we introduce Multi-Agent PRompt Optimization (MAPRO), a four-stage framework that first formulates MAS prompt optimization as a Maximum a Posteriori (MAP) inference problem and solves it using a languageguided variant of max-product belief propagation algorithm.To address credit assignment and updates the system iteratively, MAPRO employs a topology-aware refinement mechanism that integrates execution feedback and downstream blames to selectively update agent prompts.Through this process, MAPRO progressively converges to a coordinated set of agent-specific prompt policies.Across benchmarks in various tasks, MAPRO achieves stateof-the-art performance, consistently surpassing manually engineered baselines and recent automated alternatives.Beyond performance, our MAP-based formulation also delivers general guidelines for building more reliable and principled multi-agent systems in the future 1 .

Source: Findings of the Association for Computational Linguistics: EACL

DOI: 10.18653/v1/2026.findings-eacl.233

Authors: Zheyuan Zhang, Lin Ge, Hongjiang Li, Weicheng Zhu, Chuxu Zhang, Yanfang Ye

Institutions: University of Notre Dame, Amazon (United States), University of Connecticut

Language: English

Topic: Natural Language Processing Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: National Science Foundation, Division of Civil, Mechanical and Manufacturing Innovation

Awards: III: Small: A New Machine Learning Paradigm Towards Effective yet Efficient Foundation Graph Learning Models, SCC-IRG Track 1: A New AI-driven Paradigm to Promote Community Resilience for Teenagers and Young Adults in Preventing Opioid Misuse and Addiction, Collaborative Research: III: Small: Data-driven and AI-powered Innovations for Tackling the Nexus of Poverty-driven Food Insecurity and the Opioid Crisis, III: Medium: A Data-driven and AI-augmented Framework for Collaborative Decision Making to Combat Infectious Disease Outbreaks, D-ISN: An AI-augmented Framework to Detect, Disrupt, and Dismantle Opioid Trafficking Networks