ArticleNPJ digital medicine2025
Towards evaluating and building versatile large language models for medicine.
Article in NPJ digital medicine, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 28 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
28 citing papers in PubMed.
- Entity-centric evaluation of large language model responses for medical question-answering tasks.PLOS digital health · 2026Article
- Generative large language models in medicine: a scoping review of recent methodological advances.npj health systems · 2026Review
- Exploring large language models as a prescription decision support tool for rational antibiotic use: A dual-framework analysis using standardized examinations and real-world clinical cases.Exploratory research in clinical and social pharmacy · 2026Article
- Evaluating ChatGPT's Effectiveness for Arabic Dry Mouth Patient Education.Healthcare (Basel, Switzerland) · 2026Article
- Automated Review of Patient Records: Privacy-Preserving Large Language Models for Identifying Incident Nonarteritic Anterior Ischemic Optic Neuropathy at Scale.Ophthalmology science · 2026Article
- Evaluation Frameworks for Clinical Foundation Models in Specific Tasks of Unstructured Medical Text Analysis: A Scoping Review.Healthcare (Basel, Switzerland) · 2026Review
- BRIDGE: benchmarking large language models for understanding real-world clinical practice texts.Nature biomedical engineering · 2026Article
- Assessing the capability of large language models in answering pediatric critical care board-style questions.Scientific reports · 2026Article
- Automated RECIST tumor response classification through prompt-guided large language models.Scientific reports · 2026Article
- Article
- Fourier Kolmogorov-Arnold Network integrated into BioBERT-based model for Biomedical Named Entity Recognition.NPJ digital medicine · 2026Article
- Article
- Large language models for structuring knee ultrasound reports-A comparative evaluation of DeepSeek R1, gemini 2.5 flash, and GPT-4o.BMC medical informatics and decision making · 2026Article
- Article
- Holistic evaluation of large language models for medical tasks with MedHELM.Nature medicine · 2026Article
- A Survey on Medical Competence Evaluation Benchmarks for Large Language Models.Health care science · 2026Review
- Curation and Extraction of Drug-Related Entities from Reddit Platform.Proceedings. IEEE International Conference on Healthcare Informatics · 2026Article
- Virtual medicine: medical AI in human health and diseases.Military Medical Research · 2026Review
- A roadmap for medical large language models: a review of foundations, applications, and challenges.Military Medical Research · 2026Review
- Simulated patient systems powered by large language model-based AI agents offer potential for transforming medical education.Communications medicine · 2025Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
8 authors.
Funding
Abstract
In this study, we present MedS-Bench, a comprehensive benchmark to evaluate large language models (LLMs) in clinical contexts, MedS-Bench, spanning 11 high-level clinical tasks. We evaluate nine leading LLMs, e.g., MEDITRON, Llama 3, Mistral, GPT-4, Claude-3.5, etc. and found that most models struggle with these complex tasks. To address these limitations, we developed MedS-Ins, a large-scale instruction-tuning dataset for medicine. MedS-Ins comprises 58 medically oriented language corpora, totaling 5M instances with 19K instructions, across 122 tasks. To demonstrate the dataset's utility, we conducted a proof-of-concept experiment by performing instruction tuning on a lightweight, open-source medical language model. The resulting model, MMedIns-Llama 3, significantly outperformed existing models on various clinical tasks. To promote further advancements, we have made MedS-Ins fully accessible and invite the research community to contribute to its expansion. Additionally, we have launched a dynamic leaderboard for MedS-Bench, to track the development progress of medical LLMs.
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.