OpenBioRQ is a benchmark of open-ended, currently-unresolved biomedical research questions. The dataset was extracted from primary literature and clinical-trial records, refined to be self-contained, and graded by per-question rubrics. It was created by Minbyul and last updated on June 20, 2026.
Use Cases
- Evaluating AI agents on open-ended biomedical question answering based on the benchmark's rubrics.
- Training retrieval-augmented generation models for biomedical literature based on unresolved research questions.
- Benchmarking the reasoning capabilities of large language models in biomedical domains using graded questions.
Strengths
- Questions are graded by per-question rubrics, providing structured evaluation criteria.
- Data is extracted from primary literature and clinical-trial records, suggesting a foundation in real research.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Last updated 2026-06-20 09:17:56; freshness should be verified.
Provenance
- Source
- Primary literature and clinical-trial records.
- Collection Method
- Extracted and refined to be self-contained.
- Freshness
- Last updated 2026-06-20 09:17:56.