ResidencyRL: Reinforcement Learning in Simulated Clinical Environments
- URL: http://arxiv.org/abs/2608.07418v1
- Date: Fri, 07 Aug 2026 17:04:41 GMT
- Title: ResidencyRL: Reinforcement Learning in Simulated Clinical Environments
- Abstract summary: We present ResidencyRL, a reinforcement learning (RL) method for training clinical artificial intelligence (AI) agents through simulated multi-turn clinical encounters.<n>On heldout evaluations, the ResidencyRL agent improves diagnostic accuracy by 7.0% under adversarial conditions.<n>Blinded expert clinicians validated these gains, preferring the trained agent in 87.6% of side-by-side comparisons.
- Score: 53.76751756107077
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: In medical education, physicians convert academic knowledge into clinical expertise through residency: years of training across thousands of encounters, with diverse sources of feedback and progressively greater autonomy. Much of clinical reasoning relies on the patient encounter, a dialogue in which a clinician elicits history, refines diagnostic hypotheses, and decides management under uncertainty. While large language models (LLMs) excel on static medical benchmarks, methods to optimize the full sequence of clinical decisions remain underdeveloped. We present ResidencyRL, a reinforcement learning (RL) method for training clinical artificial intelligence (AI) agents through simulated multi-turn clinical encounters (up to 60 dialogue turns and 8 tool calls per trajectory). ResidencyRL pairs the policy agent with LLM simulators capable of complex, adversarial behaviors, training against a structured reward aligned to diagnostic accuracy, management quality, communication, documentation, and safety. On held-out evaluations, the ResidencyRL agent improves diagnostic accuracy by 7.0% under adversarial conditions (88.0% vs. 81.0%) and reduces missed red flag rates by 31%, demonstrating rigorous mitigation of premature closure. Blinded expert clinicians validated these gains, preferring the trained agent in 87.6% of side-by-side comparisons. The procedural competencies transfer to unseen benchmarks: the agent outperforms the base model across all six clinical axes of the AMIE multi-visit benchmark, and shows consistent directional improvements on AgentClinic and CRAFT-MD. Our findings demonstrate that sequential clinical decision-making can be effectively learned through multi-turn RL in simulation, yielding robust, generalizable capabilities, paving the way towards clinical mastery. Prospective validation with real-world workflows remains necessary to establish clinical utility.
Related papers
- Evidence-Grounded AI for Musculoskeletal Care [106.59121478639402]
OrthoPilot is a clinical artificial intelligence system powered by a large language model (LLM)<n>It integrates hospital data streams with authoritative external knowledge for continuous musculoskeletal care.<n>In a full-pathway reader study against 81 orthopaedic physicians, OrthoPilot outperformed experts with 25 years of experience.
arXiv Detail & Related papers (2026-07-14T09:03:26Z) - Benchmarking Multi-turn Medical Diagnosis: Hold, Lure, and Self-Correction [72.89352076103889]
Large language models (LLMs) achieve high accuracy in medical diagnosis when all clinical information is provided in a single turn.<n>We introduce MINT, a high-fidelity, multi-turn medical diagnosis benchmark comprising 1,035 cases.<n>We uncover three persistent behavioral patterns that significantly impact diagnostic decisions.
arXiv Detail & Related papers (2026-04-06T00:23:10Z) - Cerebra: A Multidisciplinary AI Board for Multimodal Dementia Characterization and Risk Assessment [56.62016795093786]
Cerebra is an interactive multi-agent AI team that coordinates specialized agents for EHR, clinical notes, and medical imaging analysis.<n>It supports privacy-preserving deployment by operating on structured representations and remains robust when modalities are incomplete.<n>Cerebra significantly improved expert performance, increasing accuracy by 17.5 percentage points in prospective dementia risk estimation.
arXiv Detail & Related papers (2026-03-23T05:46:45Z) - Emulating Clinician Cognition via Self-Evolving Deep Clinical Research [104.66211772795747]
We developed DxEvolve, a self-evolving diagnostic agent that bridges these gaps through an interactive deep clinical research workflow.<n>On the MIMIC-CDM benchmark, DxEvolve improved diagnostic accuracy by 11.2% on average over backbone models.<n>By transforming experience into a governable learning asset, DxEvolve supports an accountable pathway for the continual evolution of clinical AI.
arXiv Detail & Related papers (2026-03-11T11:41:51Z) - Evolving Diagnostic Agents in a Virtual Clinical Environment [75.59389103511559]
We present a framework for training large language models (LLMs) as diagnostic agents with reinforcement learning.<n>Our method acquires diagnostic strategies through interactive exploration and outcome-based feedback.<n>DiagAgent significantly outperforms 10 state-of-the-art LLMs, including DeepSeek-v3 and GPT-4o.
arXiv Detail & Related papers (2025-10-28T17:19:47Z) - MACD: Multi-Agent Clinical Diagnosis with Self-Learned Knowledge for LLM [32.0716204095227]
Large language models (LLMs) have demonstrated notable potential in medical applications.<n>This study proposes a novel Multi-Agent Clinical Diagnosis (MACD) framework.<n>It allows LLMs to self-learn clinical knowledge via a multi-agent pipeline.
arXiv Detail & Related papers (2025-09-24T12:37:11Z) - AgentClinic: a multimodal agent benchmark to evaluate AI in simulated clinical environments [2.567146936147657]
We introduce AgentClinic, a multimodal agent benchmark for evaluating large language models (LLM) in simulated clinical environments.<n>We find that solving MedQA problems in the sequential decision-making format of AgentClinic is considerably more challenging, resulting in diagnostic accuracies that can drop to below a tenth of the original accuracy.
arXiv Detail & Related papers (2024-05-13T17:38:53Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.