ConfAgents: A Conformal-Guided Multi-Agent Framework for Cost-Efficient Medical Diagnosis

TL;DR

This paper presents HealthFlow, a self-evolving AI agent for healthcare that autonomously improves its strategic planning capabilities, outperforming existing frameworks and advancing autonomous medical diagnosis.

Contribution

Introduction of HealthFlow, a novel meta-level evolution mechanism enabling AI agents to self-improve their strategic policies in healthcare tasks.

Findings

HealthFlow significantly outperforms state-of-the-art agents.

The self-evolving approach enhances strategic planning in complex tasks.

Benchmark results demonstrate improved efficiency and accuracy.

Abstract

The efficacy of AI agents in healthcare research is hindered by their reliance on static, predefined strategies. This creates a critical limitation: agents can become better tool-users but cannot learn to become better strategic planners, a crucial skill for complex domains like healthcare. We introduce HealthFlow, a self-evolving AI agent that overcomes this limitation through a novel meta-level evolution mechanism. HealthFlow autonomously refines its own high-level problem-solving policies by distilling procedural successes and failures into a durable, strategic knowledge base. To anchor our research and facilitate reproducible evaluation, we introduce EHRFlowBench, a new benchmark featuring complex, realistic health data analysis tasks derived from peer-reviewed clinical research. Our comprehensive experiments demonstrate that HealthFlow's self-evolving approach significantly outperforms state-of-the-art agent frameworks. This work marks a necessary shift from building better tool-users to designing smarter, self-evolving task-managers, paving the way for more autonomous and effective AI for scientific discovery.