BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915870322720768 |
|---|---|
| author | Koretsky, Mathew J. Willey, Maya Bianchi, Owen Alvarado, Chelsea X. Nayak, Tanay Kuznetsov, Nicole Kim, Sungwon Nalls, Mike A. Khashabi, Daniel Faghri, Faraz |
| author_facet | Koretsky, Mathew J. Willey, Maya Bianchi, Owen Alvarado, Chelsea X. Nayak, Tanay Kuznetsov, Nicole Kim, Sungwon Nalls, Mike A. Khashabi, Daniel Faghri, Faraz |
| contents | Biomedical researchers increasingly rely on large-scale structured databases for complex analytical tasks. However, current text-to-SQL systems often struggle to map qualitative scientific questions into executable SQL, particularly when implicit domain reasoning is required. We introduce BiomedSQL, the first benchmark explicitly designed to evaluate scientific reasoning in text-to-SQL generation over a real-world biomedical knowledge base. BiomedSQL comprises 68,000 question/SQL query/answer triples generated from templates and grounded in a harmonized BigQuery knowledge base that integrates gene-disease associations, causal inference from omics data, and drug approval records. Each question requires models to infer domain-specific criteria, such as genome-wide significance thresholds, effect directionality, or trial phase filtering, rather than rely on syntactic translation alone. We evaluate a range of open- and closed-source LLMs across prompting strategies and interaction paradigms. Our results reveal a substantial performance gap: Gemini-3-Pro achieves 58.1% execution accuracy, while our custom multi-step agent, BMSQL, reaches 62.6%, both well below the expert baseline of 90.0%. BiomedSQL provides a new foundation for advancing text-to-SQL systems capable of supporting scientific discovery through robust reasoning over structured biomedical knowledge bases. Our dataset is publicly available at https://huggingface.co/datasets/NIH-CARD/BiomedSQL, and our code is open-source at https://github.com/NIH-CARD/biomedsql. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_20321 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases Koretsky, Mathew J. Willey, Maya Bianchi, Owen Alvarado, Chelsea X. Nayak, Tanay Kuznetsov, Nicole Kim, Sungwon Nalls, Mike A. Khashabi, Daniel Faghri, Faraz Computation and Language Artificial Intelligence Machine Learning Biomedical researchers increasingly rely on large-scale structured databases for complex analytical tasks. However, current text-to-SQL systems often struggle to map qualitative scientific questions into executable SQL, particularly when implicit domain reasoning is required. We introduce BiomedSQL, the first benchmark explicitly designed to evaluate scientific reasoning in text-to-SQL generation over a real-world biomedical knowledge base. BiomedSQL comprises 68,000 question/SQL query/answer triples generated from templates and grounded in a harmonized BigQuery knowledge base that integrates gene-disease associations, causal inference from omics data, and drug approval records. Each question requires models to infer domain-specific criteria, such as genome-wide significance thresholds, effect directionality, or trial phase filtering, rather than rely on syntactic translation alone. We evaluate a range of open- and closed-source LLMs across prompting strategies and interaction paradigms. Our results reveal a substantial performance gap: Gemini-3-Pro achieves 58.1% execution accuracy, while our custom multi-step agent, BMSQL, reaches 62.6%, both well below the expert baseline of 90.0%. BiomedSQL provides a new foundation for advancing text-to-SQL systems capable of supporting scientific discovery through robust reasoning over structured biomedical knowledge bases. Our dataset is publicly available at https://huggingface.co/datasets/NIH-CARD/BiomedSQL, and our code is open-source at https://github.com/NIH-CARD/biomedsql. |
| title | BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2505.20321 |