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

In three linesSEMA-RAG is a multi-agent framework for retrieval-augmented generation applied to medical reasoning. It decouples the process into three specialist agents: clinical interpretation, iterative document exploration, and evidence adjudication. Tested on 5 benchmarks and 5 LLM backbones, it improves baselines by +6.46 accuracy points on average.

Summary generated by Claude — human-verified