BRIDGE: Benchmarking Large Language Models for Understanding Real-World Clinical Practice Texts

Abstract

Large language models (LLMs) are evolving rapidly and hold great promise for medical applications, yet benchmarking on real-world clinical data such as electronic health records remains limited. Most existing benchmarks rely on medical examination-style questions or PubMed-derived text, failing to capture the complexity of clinical practice, while others target specific application scenarios with limited generalizability. Here we present BRIDGE, a comprehensive multilingual benchmark comprising 87 tasks sourced from 59 real-world clinical data sources across 9 languages. It covers eight task types spanning the patient care continuum, including triage, information extraction, diagnosis, prognosis and billing coding, and involves 14 clinical specialties. We systematically evaluated 95 LLMs (including DeepSeek-R1, GPT-4o, Gemini and Qwen3) under multiple inference strategies. Results reveal substantial performance variation across model sizes, languages, natural language processing tasks and clinical specialties. Open-source LLMs can match proprietary models, while medically fine-tuned models built on older backbones often lag behind updated general-purpose LLMs. BRIDGE and its continuously updated leaderboard provide foundational resources and important references for evaluating and developing LLMs for real-world clinical text understanding.

Contribution highlights

• Here we present BRIDGE, a comprehensive multilingual benchmark comprising 87 tasks sourced from 59 real-world clinical data sources across 9 languages.
• Results reveal substantial performance variation across model sizes, languages, natural language processing tasks and clinical specialties.

Extracted from the public abstract; no claims have been added.

Code and data

Publication metadata

DOI

10.1038/s41551-026-01719-2

PMID

42310130

Version relationship

• Preprint: https://arxiv.org/abs/2504.19467
• Version of record: https://doi.org/10.1038/s41551-026-01719-2

BibTeX

@article{Wu2026Bridgebenchmarking,
  title = {BRIDGE: Benchmarking Large Language Models for Understanding Real-World Clinical Practice Texts},
  author = {Jiageng Wu and Bowen Gu and Ren Zhou and Kevin Xie and Doug Snyder and Yixing Jiang and Valentina Carducci and Richard Wyss and Rishi J Desai and Emily Alsentzer and Leo Anthony Celi and Adam Rodman and Sebastian Schneeweiss and Jonathan H Chen and Santiago Romero-Brufau and Kueiyu Joshua Lin and Jie Yang},
  journal = {Nature Biomedical Engineering},
  year = {2026},
  url = {https://www.nature.com/articles/s41551-026-01719-2},
  doi = {10.1038/s41551-026-01719-2},
  pmid = {42310130}
}