ProSEA: Problem Solving via Exploration Agents

2025-10-08

AI Summary

Existing AI agents predominantly rely on static planning, lacking adaptive reasoning and robust collaborative capabilities, which hinders their performance in complex, dynamic tasks. This paper proposes a modular multi-agent framework featuring a hierarchical “Manager-Expert” architecture. It integrates structured feedback loops with autonomous exploration mechanisms to enable dynamic task decomposition, execution-trajectory-based failure attribution, constraint-aware re-planning, and human-in-the-loop corrective refinement. The core contribution is an exploratory feedback mechanism that allows agents to explicitly articulate failure causes and newly discovered constraints, thereby significantly enhancing reasoning transparency and environmental adaptability. Evaluated on the FinanceBench benchmark, our approach substantially outperforms state-of-the-art methods, demonstrating strong robustness and efficient problem-solving capability without human intervention.

Technology Category

• Multiagent Systems: Multiagent Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Large language models (LLMs) have empowered AI agents to tackle increasingly complex tasks. However, most existing agents remain limited to static planning and brittle interactions, falling short of true collaboration or adaptive reasoning. We introduce ProSEA, a modular, general-purpose multi-agent framework designed for iterative problem solving through exploration and plan evolution. ProSEA features a hierarchical architecture in which a Manager Agent orchestrates domain-specialized Expert Agents, decomposes tasks, and adaptively replans based on structured feedback from failed attempts. Unlike prior systems, ProSEA agents report not only success or failure but also detailed reasons for failure and newly discovered constraints, enabling dynamic plan refinement informed by exploratory traces. The framework operates autonomously but supports seamless integration with human collaborators when needed. Experiments on the challenging FinanceBench benchmark demonstrate that ProSEA, even without human feedback, outperforms state-of-the-art baselines and achieves robust performance across reasoning-heavy tasks. These results underscore ProSEA's potential as a foundation for more transparent, adaptive, and human-aligned AI agents.

Problem

• Developing adaptive AI agents for complex problem solving through exploration
• Overcoming static planning limitations in current language model agents
• Enabling dynamic plan refinement via structured failure feedback mechanisms

Innovation

• Hierarchical multi-agent framework for iterative problem solving
• Dynamic plan refinement via structured failure feedback
• Autonomous operation with optional human collaboration integration