SEMA-RAG: A Self-Evolving Multi-Agent Retrieval-Augmented Generation Framework for Medical Reasoning

TL;DR

SEMA-RAG introduces a multi-agent, self-evolving retrieval-augmented generation framework that enhances medical question answering by decoupling tasks and enabling iterative reasoning, significantly improving accuracy.

Contribution

The paper presents a novel multi-agent framework with task decoupling and dynamic exploration to improve medical reasoning in retrieval-augmented generation models.

Findings

Achieves an average of +6.46 accuracy points over baselines across five benchmarks.

Effectively decouples interpretation, exploration, and adjudication tasks.

Enhances multi-stage clinical reasoning in medical question answering.

Abstract

Retrieval-Augmented Generation (RAG) is widely employed to mitigate risks such as hallucinations and knowledge obsolescence in medical question answering, yet its predominantly single-round, static retrieval paradigm misaligns with the multi-stage process of clinical reasoning. This compressed workflow induces two structural deficiencies: question-to-query translation often lacks clinically grounded semantic interpretation, and retrieval lacks iterative sufficiency feedback, making it difficult to form reliable evidence chains. We argue that both issues stem from a deeper cause: overloading a single reasoning chain with heterogeneous tasks of interpretation, exploration, and adjudication. The remedy is to reconstruct the workflow via task decoupling and dynamic multi-round exploration. To this end, we propose SEMA-RAG, a Self-Evolving Multi-Agent RAG framework for medical question answering, which assigns these roles to three specialist agents: the Interpreter Agent for clinical schema interpretation, the Explorer Agent for sufficiency-driven self-evolving retrieval, and the Arbiter Agent for evidence adjudication and answer selection. Across five benchmarks and five LLM backbones, SEMA-RAG improves the strongest baseline by +6.46 accuracy points on average, measured per backbone.