Yifan Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yifan Yang has a total of 146 co-authors. Here are the top co-authors by number of shared publications:
- Zhiyong Lu
National Library of Medicine, National Institutes of Health
7 shared publications - Qiao Jin
National Library of Medicine, National Institutes of Health, Bethesda, USA
6 shared publications Furong Huang
Department of Computer Science, University of Maryland at College Park, College Park, USA
4 shared publicationsZhiyong Lu
National Library of Medicine, National Institutes of Health, Bethesda, USA
3 shared publications- Changlin Gong
Jacob Medical Center, Albert Einstein College of Medicine, New York City, USA
2 shared publications - Guangzhi Xiong
Department of Computer Science, University of Virginia, Charlottesville, USA
2 shared publications
Latest Publications
- Open Access
- 19 September 2026
- Published by Springer Nature in Communications Medicine
Assessing safety and trustworthiness of large language models in medicine
Background: The remarkable capabilities of large language models make them increasingly compelling for use in real-world healthcare applications. However, the risks associated with using these artificial intelligence systems in medicine are not systematically understood. The aim of this study is to characterize these risks by applying five key principles for safe and trustworthy medical artificial intelligence: truthfulness, resilience, fairness, robustness, and privacy. Methods: We introduce MedGuard-Bench: a safety benchmark featuring one thousand expert-verified questions covering ten specific aspects of our five core principles. We use this comprehensive corpus to systematically evaluate sixteen commonly used large language models, assessing their safety and reliability in medical contexts. Results: We show that current large language models generally perform poorly on most of our safety tests, regardless of their safety alignment mechanisms. Our evaluation demonstrates that these models fall significantly short when compared to the high performance and reliability of human physicians. Conclusions: Despite reports indicating that advanced large language models can match or exceed human performance in various medical tasks, this study reveals a significant safety gap in current technology. This underscores the crucial need for ongoing human oversight and the implementation of strict safety guardrails before deploying these tools in clinical practice.
- 19 September 2026
- Published by Springer Nature in Communications Medicine
Open AccessAssessing safety and trustworthiness of large language models in medicine
Background: The remarkable capabilities of large language models make them increasingly compelling for use in real-world healthcare applications. However, the risks associated with using these artificial intelligence systems in medicine are not systematically understood. The aim of this study is to characterize these risks by applying five key principles for safe and trustworthy medical artificial intelligence: truthfulness, resilience, fairness, robustness, and privacy. Methods: We introduce MedGuard-Bench: a safety benchmark featuring one thousand expert-verified questions covering ten specific aspects of our five core principles. We use this comprehensive corpus to systematically evaluate sixteen commonly used large language models, assessing their safety and reliability in medical contexts. Results: We show that current large language models generally perform poorly on most of our safety tests, regardless of their safety alignment mechanisms. Our evaluation demonstrates that these models fall significantly short when compared to the high performance and reliability of human physicians. Conclusions: Despite reports indicating that advanced large language models can match or exceed human performance in various medical tasks, this study reveals a significant safety gap in current technology. This underscores the crucial need for ongoing human oversight and the implementation of strict safety guardrails before deploying these tools in clinical practice.
- 24 July 2026
- Published by Springer Nature in Nature Protocols
- p. 1-12
Tutorial: guidance on the use of large language models for medical research
- 24 July 2026
- Published by Springer Nature in Nature Protocols
- p. 1-12
Tutorial: guidance on the use of large language models for medical research
- Top Cited PapersOpen Access
- 19 June 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Memorization in large language models in medicine prevalence characteristics and implications
Large Language Models (LLMs) have demonstrated significant potential in medicine, with many studies adapting them through continued pretraining or fine-tuning on medical data. However, a key question remains: to what extent do LLMs memorize medical training data—that is, recall or regenerate content seen during continued pretraining or fine-tuning. In this work, we investigate memorization of LLMs in medicine, assessing its prevalence (frequency), characteristics (what is memorized), volume (how much), and potential downstream impacts. We systematically analyze common adaptation scenarios: (1) continued pretraining on medical corpora, (2) fine-tuning on standard medical benchmarks, and (3) fine-tuning on real-world clinical data, including over 13,000 unique inpatient records from Yale New Haven Health System. The results demonstrate that memorization is prevalent and significantly higher than that in the general domain. Memorization has distinct characteristics during continued pretraining and fine-tuning, and it is persistent: up to 87% of content memorized during continued pretraining remains after fine-tuning. Memorization can be categorized into three types: beneficial (e.g., accurate recall of clinical guidelines), uninformative (e.g., templated language), and harmful (e.g., sensitive clinical content). We offer practical recommendations to facilitate beneficial memorization, minimize uninformative memorization, and mitigate harmful memorization to protect patient privacy and improve medical utility.
- 19 June 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Top Cited PapersOpen AccessMemorization in large language models in medicine prevalence characteristics and implications
Large Language Models (LLMs) have demonstrated significant potential in medicine, with many studies adapting them through continued pretraining or fine-tuning on medical data. However, a key question remains: to what extent do LLMs memorize medical training data—that is, recall or regenerate content seen during continued pretraining or fine-tuning. In this work, we investigate memorization of LLMs in medicine, assessing its prevalence (frequency), characteristics (what is memorized), volume (how much), and potential downstream impacts. We systematically analyze common adaptation scenarios: (1) continued pretraining on medical corpora, (2) fine-tuning on standard medical benchmarks, and (3) fine-tuning on real-world clinical data, including over 13,000 unique inpatient records from Yale New Haven Health System. The results demonstrate that memorization is prevalent and significantly higher than that in the general domain. Memorization has distinct characteristics during continued pretraining and fine-tuning, and it is persistent: up to 87% of content memorized during continued pretraining remains after fine-tuning. Memorization can be categorized into three types: beneficial (e.g., accurate recall of clinical guidelines), uninformative (e.g., templated language), and harmful (e.g., sensitive clinical content). We offer practical recommendations to facilitate beneficial memorization, minimize uninformative memorization, and mitigate harmful memorization to protect patient privacy and improve medical utility.
- Open Access
- 8 June 2026
- Published by Springer Nature in Nature
- Vol. 656 (8129) , 1003-1012
GPR15-guided CD8+ T regulatory cells control intestinal inflammation
Inflammatory bowel disease (IBD) causes chronic suffering from gastrointestinal inflammation and dysfunction that can progress to colon cancer 1,2 . The prevalence of the disease is increasing, and there is an urgent need to better understand its pathogenic mechanisms to improve treatment. We show that GPR15—a G-protein-coupled receptor expressed in immune cells and described previously as an entry co-factor for human and simian immunodeficiency viruses 3 —is a marker and homing receptor for a subset of intramucosal GPR15-guided regulatory CD8 + T lymphocytes (CD8 + T IGR cells). Deleterious GPR15 gene variants in humans cause defective homing of CD8 + T IGR cells and are associated with severe early-onset IBD. Moreover, CD8 + T IGR cells are reduced in the intestinal mucosa of individuals with sporadic IBD. In mice, GPR15 deficiency impairs colonic homing of CD8 + T IGR cells, leading to accumulation of inflammatory macrophages and increased susceptibility to colitis. CD8 + T IGR cells potently kill macrophages activated by intestinal damage or disease using Fas ligand and TNF-related weak inducer of apoptosis (TWEAK). The identification of CD8 + T IGR cells yields new insights into organ-specific immune regulation and potential therapeutics for IBD.
- 8 June 2026
- Published by Springer Nature in Nature
- Vol. 656 (8129) , 1003-1012
Open AccessGPR15-guided CD8+ T regulatory cells control intestinal inflammation
Inflammatory bowel disease (IBD) causes chronic suffering from gastrointestinal inflammation and dysfunction that can progress to colon cancer 1,2 . The prevalence of the disease is increasing, and there is an urgent need to better understand its pathogenic mechanisms to improve treatment. We show that GPR15—a G-protein-coupled receptor expressed in immune cells and described previously as an entry co-factor for human and simian immunodeficiency viruses 3 —is a marker and homing receptor for a subset of intramucosal GPR15-guided regulatory CD8 + T lymphocytes (CD8 + T IGR cells). Deleterious GPR15 gene variants in humans cause defective homing of CD8 + T IGR cells and are associated with severe early-onset IBD. Moreover, CD8 + T IGR cells are reduced in the intestinal mucosa of individuals with sporadic IBD. In mice, GPR15 deficiency impairs colonic homing of CD8 + T IGR cells, leading to accumulation of inflammatory macrophages and increased susceptibility to colitis. CD8 + T IGR cells potently kill macrophages activated by intestinal damage or disease using Fas ligand and TNF-related weak inducer of apoptosis (TWEAK). The identification of CD8 + T IGR cells yields new insights into organ-specific immune regulation and potential therapeutics for IBD.
- Top Cited PapersOpen Access
- 23 October 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-11
AgentMD: Empowering language agents for risk prediction with large-scale clinical tool learning
Clinical calculators play a vital role in healthcare, but their utilization is often hindered by usability and dissemination challenges. We introduce AgentMD, a novel language agent capable of curating and applying clinical calculators across various clinical contexts. As a tool builder, AgentMD first uses PubMed to curate a diverse set of 2,164 executable clinical calculators with over 85% accuracy for quality checks and over 90% pass rate for unit tests. As a tool user, AgentMD autonomously selects and applies the relevant clinical calculators. Our evaluations show that AgentMD significantly outperforms GPT-4 for risk prediction (87.7% vs. 40.9% in accuracy). Results on 698 real-world emergency department notes confirm that AgentMD accurately computes medical risks at an individual level. Moreover, AgentMD can provide population-level insights for institutional risk management. Our study illustrates the capabilities of language agents to curate and utilize clinical calculators for both individual patient care and at-scale healthcare analytics.
- 23 October 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-11
Top Cited PapersOpen AccessAgentMD: Empowering language agents for risk prediction with large-scale clinical tool learning
Clinical calculators play a vital role in healthcare, but their utilization is often hindered by usability and dissemination challenges. We introduce AgentMD, a novel language agent capable of curating and applying clinical calculators across various clinical contexts. As a tool builder, AgentMD first uses PubMed to curate a diverse set of 2,164 executable clinical calculators with over 85% accuracy for quality checks and over 90% pass rate for unit tests. As a tool user, AgentMD autonomously selects and applies the relevant clinical calculators. Our evaluations show that AgentMD significantly outperforms GPT-4 for risk prediction (87.7% vs. 40.9% in accuracy). Results on 698 real-world emergency department notes confirm that AgentMD accurately computes medical risks at an individual level. Moreover, AgentMD can provide population-level insights for institutional risk management. Our study illustrates the capabilities of language agents to curate and utilize clinical calculators for both individual patient care and at-scale healthcare analytics.
- Open Access
- 9 October 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-10
Adversarial prompt and fine-tuning attacks threaten medical large language models
The integration of Large Language Models (LLMs) into healthcare applications offers promising advancements in medical diagnostics, treatment recommendations, and patient care. However, the susceptibility of LLMs to adversarial attacks poses a significant threat, potentially leading to harmful outcomes in delicate medical contexts. This study investigates the vulnerability of LLMs to two types of adversarial attacks–prompt injections with malicious instructions and fine-tuning with poisoned samples–across three medical tasks: disease prevention, diagnosis, and treatment. Utilizing real-world patient data, we demonstrate that both open-source and proprietary LLMs are vulnerable to malicious manipulation across multiple tasks. We discover that while integrating poisoned data does not markedly degrade overall model performance on medical benchmarks, it can lead to noticeable shifts in fine-tuned model weights, suggesting a potential pathway for detecting and countering model attacks. This research highlights the urgent need for robust security measures and the development of defensive mechanisms to safeguard LLMs in medical applications, to ensure their safe and effective deployment in healthcare settings.
- 9 October 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-10
Open AccessAdversarial prompt and fine-tuning attacks threaten medical large language models
The integration of Large Language Models (LLMs) into healthcare applications offers promising advancements in medical diagnostics, treatment recommendations, and patient care. However, the susceptibility of LLMs to adversarial attacks poses a significant threat, potentially leading to harmful outcomes in delicate medical contexts. This study investigates the vulnerability of LLMs to two types of adversarial attacks–prompt injections with malicious instructions and fine-tuning with poisoned samples–across three medical tasks: disease prevention, diagnosis, and treatment. Utilizing real-world patient data, we demonstrate that both open-source and proprietary LLMs are vulnerable to malicious manipulation across multiple tasks. We discover that while integrating poisoned data does not markedly degrade overall model performance on medical benchmarks, it can lead to noticeable shifts in fine-tuned model weights, suggesting a potential pathway for detecting and countering model attacks. This research highlights the urgent need for robust security measures and the development of defensive mechanisms to safeguard LLMs in medical applications, to ensure their safe and effective deployment in healthcare settings.
- Top Cited PapersOpen Access
- 18 November 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-14
Matching patients to clinical trials with large language models
Patient recruitment is challenging for clinical trials. We introduce TrialGPT, an end-to-end framework for zero-shot patient-to-trial matching with large language models. TrialGPT comprises three modules: it first performs large-scale filtering to retrieve candidate trials (TrialGPT-Retrieval); then predicts criterion-level patient eligibility (TrialGPT-Matching); and finally generates trial-level scores (TrialGPT-Ranking). We evaluate TrialGPT on three cohorts of 183 synthetic patients with over 75,000 trial annotations. TrialGPT-Retrieval can recall over 90% of relevant trials using less than 6% of the initial collection. Manual evaluations on 1015 patient-criterion pairs show that TrialGPT-Matching achieves an accuracy of 87.3% with faithful explanations, close to the expert performance. The TrialGPT-Ranking scores are highly correlated with human judgments and outperform the best-competing models by 43.8% in ranking and excluding trials. Furthermore, our user study reveals that TrialGPT can reduce the screening time by 42.6% in patient recruitment. Overall, these results have demonstrated promising opportunities for patient-to-trial matching with TrialGPT.
- 18 November 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-14
Top Cited PapersOpen AccessMatching patients to clinical trials with large language models
Patient recruitment is challenging for clinical trials. We introduce TrialGPT, an end-to-end framework for zero-shot patient-to-trial matching with large language models. TrialGPT comprises three modules: it first performs large-scale filtering to retrieve candidate trials (TrialGPT-Retrieval); then predicts criterion-level patient eligibility (TrialGPT-Matching); and finally generates trial-level scores (TrialGPT-Ranking). We evaluate TrialGPT on three cohorts of 183 synthetic patients with over 75,000 trial annotations. TrialGPT-Retrieval can recall over 90% of relevant trials using less than 6% of the initial collection. Manual evaluations on 1015 patient-criterion pairs show that TrialGPT-Matching achieves an accuracy of 87.3% with faithful explanations, close to the expert performance. The TrialGPT-Ranking scores are highly correlated with human judgments and outperform the best-competing models by 43.8% in ranking and excluding trials. Furthermore, our user study reveals that TrialGPT can reduce the screening time by 42.6% in patient recruitment. Overall, these results have demonstrated promising opportunities for patient-to-trial matching with TrialGPT.
- Top Cited PapersOpen Access
- 10 September 2024
- Published by Springer Nature in Communications Medicine
- Vol. 4 (1) , 1-6
Unmasking and quantifying racial bias of large language models in medical report generation
Background: Large language models like GPT-3.5-turbo and GPT-4 hold promise for healthcare professionals, but they may inadvertently inherit biases during their training, potentially affecting their utility in medical applications. Despite few attempts in the past, the precise impact and extent of these biases remain uncertain. Methods: We use LLMs to generate responses that predict hospitalization, cost and mortality based on real patient cases. We manually examine the generated responses to identify biases. Results: We find that these models tend to project higher costs and longer hospitalizations for white populations and exhibit optimistic views in challenging medical scenarios with much higher survival rates. These biases, which mirror real-world healthcare disparities, are evident in the generation of patient backgrounds, the association of specific diseases with certain racial and ethnic groups, and disparities in treatment recommendations, etc. Conclusions: Our findings underscore the critical need for future research to address and mitigate biases in language models, especially in critical healthcare applications, to ensure fair and accurate outcomes for all patients.
- 10 September 2024
- Published by Springer Nature in Communications Medicine
- Vol. 4 (1) , 1-6
Top Cited PapersOpen AccessUnmasking and quantifying racial bias of large language models in medical report generation
Background: Large language models like GPT-3.5-turbo and GPT-4 hold promise for healthcare professionals, but they may inadvertently inherit biases during their training, potentially affecting their utility in medical applications. Despite few attempts in the past, the precise impact and extent of these biases remain uncertain. Methods: We use LLMs to generate responses that predict hospitalization, cost and mortality based on real patient cases. We manually examine the generated responses to identify biases. Results: We find that these models tend to project higher costs and longer hospitalizations for white populations and exhibit optimistic views in challenging medical scenarios with much higher survival rates. These biases, which mirror real-world healthcare disparities, are evident in the generation of patient backgrounds, the association of specific diseases with certain racial and ethnic groups, and disparities in treatment recommendations, etc. Conclusions: Our findings underscore the critical need for future research to address and mitigate biases in language models, especially in critical healthcare applications, to ensure fair and accurate outcomes for all patients.
- Top Cited PapersOpen Access
- 25 April 2024
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 154, 104646
A survey of recent methods for addressing AI fairness and bias in biomedicine
- 25 April 2024
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 154, 104646
Top Cited PapersOpen AccessA survey of recent methods for addressing AI fairness and bias in biomedicine
- Open Access
- 4 April 2024
- Published by Elsevier in Journal of Cycling and Micromobility Research
- Vol. 2
Comparing e-bike and conventional bicycle use patterns in a public bike share system: A case study of Richmond, VA
- 4 April 2024
- Published by Elsevier in Journal of Cycling and Micromobility Research
- Vol. 2
Open AccessComparing e-bike and conventional bicycle use patterns in a public bike share system: A case study of Richmond, VA