PubMed Reasoner: Dynamic Reasoning-based Retrieval for Evidence-Grounded Biomedical Question Answering
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911550506270720 |
|---|---|
| author | Zhang, Yiqing Liu, Xiaozhong Murai, Fabricio |
| author_facet | Zhang, Yiqing Liu, Xiaozhong Murai, Fabricio |
| contents | Trustworthy biomedical question answering (QA) systems must not only provide accurate answers but also justify them with current, verifiable evidence. Retrieval-augmented approaches partially address this gap but lack mechanisms to iteratively refine poor queries, whereas self-reflection methods kick in only after full retrieval is completed. In this context, we introduce PubMed Reasoner, a biomedical QA agent composed of three stages: self-critic query refinement evaluates MeSH terms for coverage, alignment, and redundancy to enhance PubMed queries based on partial (metadata) retrieval; reflective retrieval processes articles in batches until sufficient evidence is gathered; and evidence-grounded response generation produces answers with explicit citations. PubMed Reasoner with a GPT-4o backbone achieves 78.32% accuracy on PubMedQA, slightly surpassing human experts, and showing consistent gains on MMLU Clinical Knowledge. Moreover, LLM-as-judge evaluations prefer our responses across: reasoning soundness, evidence grounding, clinical relevance, and trustworthiness. By orchestrating retrieval-first reasoning over authoritative sources, our approach provides practical assistance to clinicians and biomedical researchers while controlling compute and token costs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_27335 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PubMed Reasoner: Dynamic Reasoning-based Retrieval for Evidence-Grounded Biomedical Question Answering Zhang, Yiqing Liu, Xiaozhong Murai, Fabricio Computation and Language I.2.7; H.3.3; I.2.1; J.3 Trustworthy biomedical question answering (QA) systems must not only provide accurate answers but also justify them with current, verifiable evidence. Retrieval-augmented approaches partially address this gap but lack mechanisms to iteratively refine poor queries, whereas self-reflection methods kick in only after full retrieval is completed. In this context, we introduce PubMed Reasoner, a biomedical QA agent composed of three stages: self-critic query refinement evaluates MeSH terms for coverage, alignment, and redundancy to enhance PubMed queries based on partial (metadata) retrieval; reflective retrieval processes articles in batches until sufficient evidence is gathered; and evidence-grounded response generation produces answers with explicit citations. PubMed Reasoner with a GPT-4o backbone achieves 78.32% accuracy on PubMedQA, slightly surpassing human experts, and showing consistent gains on MMLU Clinical Knowledge. Moreover, LLM-as-judge evaluations prefer our responses across: reasoning soundness, evidence grounding, clinical relevance, and trustworthiness. By orchestrating retrieval-first reasoning over authoritative sources, our approach provides practical assistance to clinicians and biomedical researchers while controlling compute and token costs. |
| title | PubMed Reasoner: Dynamic Reasoning-based Retrieval for Evidence-Grounded Biomedical Question Answering |
| topic | Computation and Language I.2.7; H.3.3; I.2.1; J.3 |
| url | https://arxiv.org/abs/2603.27335 |