Condition-Gated Reasoning for Context-Dependent Biomedical Question Answering

TL;DR

This work tackles the gap in biomedical QA where clinical decisions are contingent on patient-specific factors. It introduces CondMedQA, a benchmark designed to test conditional multi-hop reasoning, and Condition-Gated Reasoning (CGR), a framework that builds condition-aware knowledge graphs and gates reasoning paths to ensure contextually valid inferences. CGR extracts condition-rich n-tuples, normalizes entities with UMLS, and performs gated graph traversal guided by a single LLM-based condition evaluation, yielding grounded, path-traced answers. Across CondMedQA and existing biomedical QA benchmarks, CGR achieves state-of-the-art performance on condition-sensitive questions while maintaining competitive results on factual tasks, highlighting the importance of explicit conditional modeling for robust medical reasoning and interpretability.

Abstract

Current biomedical question answering (QA) systems often assume that medical knowledge applies uniformly, yet real-world clinical reasoning is inherently conditional: nearly every decision depends on patient-specific factors such as comorbidities and contraindications. Existing benchmarks do not evaluate such conditional reasoning, and retrieval-augmented or graph-based methods lack explicit mechanisms to ensure that retrieved knowledge is applicable to given context. To address this gap, we propose CondMedQA, the first benchmark for conditional biomedical QA, consisting of multi-hop questions whose answers vary with patient conditions. Furthermore, we propose Condition-Gated Reasoning (CGR), a novel framework that constructs condition-aware knowledge graphs and selectively activates or prunes reasoning paths based on query conditions. Our findings show that CGR more reliably selects condition-appropriate answers while matching or exceeding state-of-the-art performance on biomedical QA benchmarks, highlighting the importance of explicitly modeling conditionality for robust medical reasoning.

Figures (3)

• Figure 1: Example of condition-gated reasoning in the biomedical domain. Existing KG-RAG extracts triples and retrieves contraindicated treatments. CGR extracts n-tuples with patient-specific conditions, gating unsafe paths and retrieving only safe alternatives.
• Figure 2: Overview of Condition-Gated Reasoning (CGR) compared to traditional KG-RAG. Given a clinical query about a patient with hypertension and bilateral renal artery stenosis (BRAS), traditional KG-RAG extracts standard relation triples and traverses all paths indiscriminately, retrieving evidence for contraindicated treatments (e.g., Lisinopril, an ACE inhibitor). CGR extends triples to n-tuples that include patient-specific conditions as gating constraints (e.g., $\neg$BRAS, $\neg$pregnancy), masking contraindicated paths during graph traversal and assembling only condition-appropriate evidence, yielding the correct answer (Amlodipine).
• Figure 3: Hyperparameter sensitivity analysis. Top row: varying $k_{\text{paths}} \in \{1, 3, 5, 10\}$. Bottom row: varying $k_{\text{nodes}} \in \{1, 3, 5, 10\}$. Based on these results, we set $k_{\text{paths}} = 3$ and $k_{\text{nodes}} = 5$ for all experiments.