Reproduce “Propagating construction-time knowledge quality into medical question answering: A framework grounded in clinical guidelines” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jie Hu, Junjie Wang, Shan Lu, Yifang Hu, Gong Cheng, Yun Liu
The authors present a question-answering system that carries construction-time quality scores and source links for retained knowledge-graph triples into answer-time retrieval and generation. It combines structural conformance and evidential support to rank evidence and guide how cautiously the generator presents it. In Chinese diabetes-guideline QA, omissions fell from 16.3% to 5.3% and conflicts from 16.3% to 2.7%; findings are limited to this setting.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.