FindLink improves assessing connections among scientific findings
FindLink: Learning to Assess Scientific Connections from Source-Grounded Premises
Computation and Language
Summary
Scientific discoveries often come from many studies with different conditions, making it hard to see how their findings connect. The authors created FindLink, a method that first finds key study details and then puts them together to figure out supported relationships and remaining questions. They tested this on thousands of cancer research questions and showed it answers better than simpler approaches. Their work shows that understanding scientific ideas needs both finding evidence and seeing how that evidence links up.
What this means in practice
- •For clinical data teams: Automatically evaluate clinical study findings for consistency and support in cancer research reports to assist medical evidence synthesis.
- •For biomedical knowledge engineers: Build tools that trace scientific claims to original studies and organize them to evaluate hypothesis support and knowledge relations.
Authors
Jingxuan Wei, Song Tan, Xiaohan Liu, Chao Sang, Changwei Wang, Linzhuang Sun, Bihui Yu, Fuqiang Wang, Junnan Zhu
Abstract
Scientific discovery from the literature requires assessing connections among findings obtained under different study conditions. Source attribution makes individual findings traceable, but does not determine what they jointly support. We introduce GI-TraceBench, a benchmark of 10,282 questions in gastrointestinal oncology that unifies provenance question answering, knowledge discovery, and hypothesis evaluation through shared Paper, Claim, and Evidence Span records. We further propose FindLink, a two-stage framework for provenance-grounded scientific reasoning. Premise Grounding constructs source-bound premises with explicit study context and support status. Structured Reasoning organizes these premises into support plans that represent both supported relations and unresolved conditions, then generates evidence-conditioned answers using deterministic compiler feedback. The compiler verifies structural and source-binding consistency, but does not assess scientific entailment. On the 1,545-question fixed-evidence test set, FindLink improves the balanced family macro score from 67.95 to 87.25 over direct supervised fine-tuning of the same backbone. A retrained variant without Structured Reasoning achieves slightly higher premise-identifier F1, but substantially lower relation and evidence-sufficiency F1. These results indicate that scientific reasoning requires not only identifying grounded premises, but also assessing what connections those premises support.