Metraipolis › archive › dc_ea99bd1b2014

EHR-RobustGym: Benchmarking and Training Agents for Robust Clinical Reasoning — arXiv:2609.39371

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.39371
License (as archived): CC0-1.0
Archived: 2026-10-07T12:42:43.196Z
SHA-256 of the archived text: f03f7f5589492c3f9fa4e7ae006c715842ae95a111030b8176cc0ab88ed632c6
Stable URL of this page: https://metraipolis.com/documento/dc_ea99bd1b2014

How to cite

EHR-RobustGym: Benchmarking and Training Agents for Robust Clinical Reasoning — arXiv:2609.39371 — arXiv (en), https://arxiv.org/abs/2609.39371 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ea99bd1b2014 (Metraipolis archive dc_ea99bd1b2014, sha256 f03f7f5589492c3f9fa4e7ae006c715842ae95a111030b8176cc0ab88ed632c6)

BibTeX

@misc{dcea99bd1b2014,
  title = {EHR-RobustGym: Benchmarking and Training Agents for Robust Clinical Reasoning — arXiv:2609.39371},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39371},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ea99bd1b2014; sha256 f03f7f5589492c3f9fa4e7ae006c715842ae95a111030b8176cc0ab88ed632c6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39371. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In hospital workflows, electronic health records (EHRs) are often noisy, and may not contain the evidence needed to confirm events or measurements referenced in a clinical query. Even when database retrieval succeeds, clinical agents can overlook such discrepancies and return plausible but unsupported answers. We introduce EHR-RobustGym, a scalable and interactive environment for evaluating and training robust clinical agents grounded in noisy EHRs. Built on MIMIC-IV hospital records (365K patients, 31 tables, and over 500M records), EHR-RobustGym comprises 5,486 Clean-Noise pairs spanning six clinical intents and both patient-level and population-level queries. The pairs test robustness to Record-level, Value-level, and Query-level noise, while interactive SQL/Python execution and outcome verification support trajectory collection and training. Evaluating multiple LLMs reveals substantial robustness gaps: average task success across proprietary and large-scale open-weight models drops from 62.2% on Clean questions to 37.9% on Noise questions. At k=4, pass^k consistency falls below 50% for most evaluated models, exposing instability in clinical task completion. Supervised fine-tuning and reinforcement learning in EHR-RobustGym improve performance, with gains generalizing to five external EHR benchmarks. Together, these results position EHR-RobustGym as a testbed for evaluating and improving the evidence-grounded robustness of clinical agents.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.39371 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T12:14:13.758Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).