A self-correcting Agentic Graph RAG for clinical decision support in hepatology
Yalan Hu, Wenjie Xuan, Qingqing Zhou, Zhi Li, Ya Li, Jili Hu, Fang Fang · 2025
Read A self-correcting Agentic Graph RAG for clinical decision support in hepatology by Yalan Hu, Wenjie Xuan, Qingqing Zhou, Zhi Li, Ya Li, Jili Hu, Fang Fang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12748213.txt.
Description
Yalan Hu and co-authors present a self correcting graph retrieval system for liver medicine decision support in a 2025 paper. Clinical knowledge grows fast, and language models can hallucinate. Standard retrieval systems also struggle to use complex medical knowledge structures, such as verified knowledge graphs.
Their framework runs a retrieve, evaluate, refine loop. Agents generate graph searches, validate them semantically, and improve them step by step to build an accurate context for the answer. On a custom set of clinical questions, the system beat strong baselines on faithfulness, context recall, and answer relevance, with scores above nine tenths.
Catalog record
| CREATOR | Yalan Hu, Wenjie Xuan, Qingqing Zhou, Zhi Li, Ya Li, Jili Hu, Fang Fang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |