AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

GALAX: Graph-Augmented Language Model for Explainable Reinforcement-Guided Subgraph Reasoning in Precision Medicine

Heming Zhang, Di Huang, Wenyu Li, Michael A Province, Yixin Chen, Philip Payne, Fuhai Li

ICLR 2026 · Large language models › Retrieval and knowledge · approach: Domain-specific QA datasets · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
task-adaptive finetuning on Target-QA instead of no finetuningLLaMA3-8B + Omics · Target-QA (TCGA, overall)precision: 0.0071 -> 0.5250Improved
domain finetuning on biomedical text instead of noneLLaMA3-8B + Omics · Target-QA (TCGA, overall)precision: 0.0071 -> 0.0179Improved
add static KG to task-finetuned LLML3-FT(QA) + Omics · Target-QA (TCGA, overall)precision: 0.5250 -> 0.5185No clear effect
use GALAX reinforcement-guided subgraph reasoning instead of static KGL3-FT(QA) + Omics + KG · Target-QA (TCGA, overall)precision: 0.5185 -> 0.5472Improved
use GALAX instead of GNN-RAGgraph RAG baseline · Target-QA (TCGA, overall)recall: 0.4735 -> 0.5332Improved
replace LLaMA3-8B backbone with Qwen2.5-7BGALAX · Target-QA (TCGA, overall)precision: 0.5472 -> 0.5445No clear effect
add epigenomic data input (ablation)GALAX · Target-QA (TCGA, overall)precision: 0.5163 -> 0.5472Improved
add genomic data input (ablation)GALAX · Target-QA (TCGA, overall)precision: 0.4277 -> 0.5472Improved
add transcriptomic data input (ablation)GALAX · Target-QA (TCGA, overall)precision: 0.4065 -> 0.5472Improved
add proteomic data input (ablation)GALAX · Target-QA (TCGA, overall)precision: 0.3980 -> 0.5472Improved
keep all omics instead of removing all (ablation)GALAX · Target-QA (TCGA, overall)precision: 0.3793 -> 0.5472Improved
keep all edges instead of dropping 20% of KG edges (ablation)GALAX · Target-QA (TCGA, overall)recall: 0.5061 -> 0.5332Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy