AI for clinical decision support, patient records, medical education, and language models in healthcare.
92 papers in this topic (ICML 2026).
OpenEvidence requires a US NPI and left the EU/UK. An alternative for clinicians elsewhere: evidence-tiered answers, cite-or-abstain, no ads. Verified July 2026.
The Phở team
Read more →Measured evidence: language models fabricate citations at scale, and fail nearly half the time when retrieving medical literature. Here is how Phở does it differently.
The Phở team
Read more →Zhitao He, Haolin Yang, Zeyu Qin, Yi R. Fung
While Large Language Models (LLMs) have achieved remarkable success in dyadic (one-on-one) instruction, they face significant challenges in One-to-Many alignment, such as clinical…
Tingting Chen, Beibei Lin, Zifeng Yuan, Qiran Zou +4
Many scientific problems are underdetermined: multiple distinct hypotheses are equally consistent with the same observations. In such settings, effective inference requires not…
Dian Ding, Liren Dong, Yu Lu, Juntao Zhou +4
Gastrointestinal (GI) motility assessment via bowel sounds (BS) offers a non-invasive alternative to resource-intensive clinical standards. However, the diagnostic utility of BS…
Lina Zhang, Tonmoy Monsoor, Peizheng Li, Jiarui Cui +4
While Multimodal Large Language Models (MLLMs) have demonstrated remarkable proficiency in general video understanding, their capacity to interpret involuntary, and…
Zongzhe Xu, Zitao Shuai, Eideen Mozaffari, Ravi Shankar Aysola +2
We present SleepLM, a family of sleep-language foundation models that enable human sleep alignment, interpretation, and interaction with natural language. Despite the critical…
Marie Brockschmidt, Maresa Schröder, Stefan Feuerriegel
Survival analysis is a cornerstone of clinical research by modeling time-to-event outcomes such as metastasis, disease relapse, or patient death. Unlike standard tabular data,…
Aleix Segui Ugalde, Wesley Armour
In AI for Science, physics-informed losses are increasingly used to train learned compressors for scientific data, but their rate--distortion implications remain poorly…
Heng Rao, Jason Zipeng Zhang, Yu Gu, Zhenghao Liu +4
Predicting long-horizon trajectories of biological dynamical systems remains challenging due to substantial system heterogeneity. Most existing machine learning approaches are…
Rituparna Datta, Zihan Guan, Baltazar Espinoza, Yiqi Su +4
Epidemic modeling is essential for public health planning, yet traditional approaches rely on fixed model classes that require manual redesign as pathogens, policies, and scenario…
Luke Nightingale, Joseph Tuersley, Scott Warchal, Andrea Cairoli +4
Phenotypic screening experiments produce many microscope images of cells under diverse perturbations, with biologically significant responses often subtle or difficult to identify…
Silas Ruhrberg Estévez, Christopher Chiu, Mihaela van der Schaar
Modern clinical practice relies on evidence-based guidelines implemented as compact scoring systems composed of a small number of interpretable decision rules. While…
Ziyu Zhao, Yiyang Liu, Yajiao Wang, Xiaotao Wang +4
Despite progress of Multimodal Large Language Models (MLLMs) in biomedical visual question answering (VQA), existing benchmarks provide limited assessment of their scientific…
Yuting Yan, Yinghao Fu, Wendi Ren, Haozhou Gao +1
Diagnosing rare diseases remains a persistent challenge, often hindered by cognitive anchoring: once clinicians settle on a common diagnosis, they often discount alternative…
Dionizije Fa, Marko Čuljak, Bruno Pandža, Mateo Čupić
We introduce BioAgent Bench, an evaluation suite designed for measuring the performance and robustness of AI agents in common bioinformatics tasks. The suite consists of manually…
Yuyang Liu, Liuzhenghao Lv, Xiancheng Zhang, Jingya Wang +2
The realization of autonomous scientific experimentation is currently limited by LLMs' struggle to grasp the strict procedural logic and accuracy required by biological protocols.…
Guanghui Min, Tianhao Huang, Ke Wan, Qi R. Wang +1
Reliable epidemic forecasting is critical for public health decision-making yet remains challenging due to data sparsity and the non-stationary nature of disease dynamics. While…
Silas Ruhrberg Estévez, Nicolas Huynh, Tennison Liu, Roderik M. Kortlever +3
Inferring dynamics from population snapshots is a fundamental challenge in machine learning and biology. In scRNA-sequencing (scRNA-seq), destructive measurements preclude direct…
Akash Ghosh, Srivarshinee Sridhar, Raghav Kaushik Ravi, Muhsin Muhsin +2
Integrating language models (LMs) in healthcare systems holds great promise for improving medical workflows and decision-making. However, a critical barrier to their global…
Chi-Min Chan, Ehsan Hajiramezanali, Xiner Li, Edward De Brouwer +4
In scientific reasoning tasks, the veracity of the reasoning process is as critical as the final outcome. While Process Reward Models (PRMs) offer a solution to the coarse-grained…
Ivan Svatko, Maxime Sanchez, Ihab Bendidi, Gilles Cottrell +1
Representation learning has driven major advances in natural image analysis by enabling models to acquire high-level semantic features. In microscopy imaging, however, it remains…
Xinyu Pang, Zhanke Zhou, Xuan Li, Fangrui Lv +4
Symbolic regression (SR) discovers compact mathematical expressions from data, yet recent LLM-based evolutionary methods remain sample-inefficient because they rely mainly on…
Shaoqing Duan, Haofei Song, Xintian Mao, Qingli Li +1
Defocus deblurring in pathological microscopy remains challenging due to the spatially varying and locally discontinuous nature of optical blur induced by a position-dependent…
Bowen Shi, Weiwei Cao, Ruifeng Yuan, Wanxing Chang +4
Vision–language pre-training (VLP) holds great promise for general-purpose medical AI by leveraging radiology reports as rich textual supervision, yet existing methods struggle…
Changshuo Liu, Wu Junran, Zhongle Xie, Wenqiao Zhang +4
Generative AI is reshaping healthcare, yet most existing advances rely on hospital-grade devices, which limits their accessibility and potential for health management outside…
Yucheng Xing, Ling Huang, Jingying Ma, Ruping Hong +4
Whole-slide images (WSIs) are widely used for cancer survival analysis because of their comprehensive histopathological information at both cellular and tissue levels, enabling…
Yuhan Wang, Yuanyuan Zou, Jie Cheng, Bin Dai +1
Accurately capturing local variations in long series has always been one of the most challenging problems in time-series forecasting especially in medical signals, where local…
Jun Li, Mingxuan Liu, Jiazhen Pan, Che Liu +3
Clinical abnormality grounding for rare diseases is often hindered by data scarcity, rendering supervised fine-tuning infeasible and single-pass inference highly unstable. Thus,…
Jiarui Jin, Haoyu Wang, Xingliang Wu, Xiaocheng Fang +4
Electrocardiography (ECG) serves as an indispensable diagnostic tool in clinical practice, yet existing multimodal large language models (MLLMs) remain unreliable for ECG…
Wei Xiong, Jiangtong Li, Jie Li, Kun Zhu +1
Electroencephalography foundation models (EEG-FMs) have advanced brain signal analysis, but the lack of standardized evaluation benchmarks impedes model comparison and scientific…
McClain Thiel, Angus G. Cunningham, Chris P Barnes
We compare the efficacy and distributional effects of supervised fine-tuning (SFT) and reinforcement learning (RL) post-training for PlasmidGPT, a foundation model for…
Chenyu Lian, Hong-Yu Zhou, Jing Qin
Disease screening is critical for early detection and timely intervention in clinical practice. However, most current screening models for medical images suffer from limited…
Dongkyu Cho, Miao Zhang, Gregory D Lyng, Rumi Chunara
Data augmentation is a widely used strategy to improve model robustness and generalization by enriching training datasets with synthetic examples. While large language models…
Pengfei Hu, Chang Lu, Feifan Liu, Yue Ning
Deep learning models for clinical event prediction on electronic health records (EHR) often suffer performance degradation when deployed under different data distributions. While…
Yingyu Chen, Yongqiang Huang, Yang Qin, Ziyuan Yang +3
In clinical practice, patients often present with multiple co-occurring diseases, yet most existing Multi-Label-Diagnosis (MLD) methods treat diagnosis as a rigid discriminative…
Yinbo Liu, Keyang Ye, Wenshan Sun, Handi Gao +1
Reconstructing unified continuous dynamics from sparse, non-contiguous, and unpaired point cloud snapshots remains a fundamental challenge in spatiotemporal analysis for computer…
Wenhao Wu, Zhentao Tang, Yafu Li, Shixiong Kai +4
Large Language Models (LLMs) exhibit high reasoning capacity in medical question-answering, but their tendency to produce hallucinations and outdated knowledge poses critical…
Zhaokun Yan, Shan Xu, Wuzheng Dong, Zhaohan Liu +4
Public health reasoning requires population level inference grounded in scientific evidence, expert consensus, and safety constraints. However, it remains underexplored as a…
Mingcheng Zhu, Zhiyao Luo, Yu Liu, Tingting Zhu
By processing electronic health records (EHRs) as natural language sequences, large language models (LLMs) have shown potential in clinical prediction tasks such as mortality…
Jiaqi Men, Hua Liu, Yiming Tang, Jinhong You +2
Accurate survival prediction in kidney transplantation is critical yet challenging due to the complex interplay between functional biomarkers and patient characteristics under…
Yichi Zhang, Nabeel Seedat, Yinpeng Dong, Peng Cui +2
As LLM-powered agents have been used for high-stakes decision-making, such as clinical diagnosis, it becomes critical to develop reliable verification of their decisions to…
Zelin Zang, WenZhe Li, Yongjie Xu, Chang Yu +4
In single-cell research, tracing and analyzing high-throughput single-cell differentiation trajectories is crucial for understanding biological processes. Key to this is the…
Sirui Li, Shuhan Xiao, Mihir Joshi, Ahmed Metwally +3
The rise of large language models (LLMs) has shifted time series analysis from narrow analytics to general-purpose reasoning. Yet, existing benchmarks cover only a small set of…
Yiqi Su, Ray Lee, Jiaming Cui, Naren Ramakrishnan
Epidemiological forecasting from surveillance data is a hard problem and hybridizing mechanistic compartmental models with neural models is a natural direction. The mechanistic…
Maria Elkjær Montgomery, Christian Igel, Mikkel Fruelund Odgaard, Martin Sillesen +1
How do we encode numeric values in transformer-based sequence processing, particularly in electronic health record (EHR) data? We systematically compare discrete, continuous, and…
Shuo Tang, Jiadong Zhang, Gengxian Zhou, Qizhao Jin +4
While deep learning-based weather forecasting paradigms have made significant strides, addressing extreme weather diagnostics remains a formidable challenge. This gap exists…
Yang Song, Yixuan Zhang, Lingfa Meng, Tongyuan Hu +4
Parameter-efficient adaptation has made LLMs practical for domain prediction, but standard LoRA still relies on a static low-rank update and does not expose the latent…
Pengkai Wang, Pengwei Liu, Qi Zuo, Zhijie Sang +2
Reinforcement learning (RL) has powered many recent breakthroughs in large language models (LLMs), especially for tasks where rewards can be computed automatically, such as code…
Evgeny Saveliev, Samuel Holt, Nabeel Seedat, David L. Bentley +2
Large Language Models (LLMs) offer a promising avenue for scientific discovery, yet their application to symbolic regression is often constrained by inefficient search strategies…
Akash Pandey, Wei Chen, Sinan Keten
Designing biological sequences such as proteins and DNA for desired properties is challenging due to vast search spaces and limited wet lab evaluation budgets. Current…
Huayu Li, ZhengXiao He, Xiwen Chen, Jingjing Wang +3
Learning meaningful representations from medical time series (MedTS), such as ECG or EEG signals, is a critical challenge. These signals are often high-dimensional,…
Sofia Ek, Dave Zachariah
Learning beneficial treatment allocations for a patient population is an important problem in precision medicine. For such allocations, a certain proportion of treated patients…
Huimin Yan, Liang Bai, Xian Yang, Long Chen
Most existing CLIP-style medical vision--language pretraining methods rely on global or local alignment with substantial paired data. However, global alignment is easily dominated…
Yihang Liu, Longzhen Yang, Jiaxiong Yang, Ying Wen +2
Medical foundation models (MFMs) aim to learn universal representations from multimodal medical images that can generalize effectively to diverse downstream clinical tasks.…
Ziquan Wei, Tingting Dan, Guorong Wu
Despite the central role of sensor-derived measurements such as imaging traits and plasma biomarkers in biomedical research and clinical practice, existing generative models for…
Anglin Liu, Ruichao Chen, Yi Lu, Hongxia Xu +1
Despite recent Multimodal Large Language Models (MLLMs)' linguistic prowess in medical diagnosis, we find even state-of-the-art MLLMs suffer from a critical perceptual deficit:…
Tianbo Wang, Yuqing Ma, Lingyan Meng, Zhange Zhang +4
Medical Large Vision-Language Models (Med-LVLMs) suffer from severe hallucinations, posing critical safety risks in clinical deployment. Editing LVLM activations has shown promise…
Yu Zhao, Hao Guan, Yongcheng Jing, Ying Zhang +1
Large Language Models (LLMs) have shown strong potential in complex medical reasoning yet face diminishing gains under inference scaling laws. While existing studies augment LLMs…
Harshit Rajgarhia, Shuubham Ojha, Asif Shaik, Akhil Pothanapalli +3
Medical audio data is difficult to collect due to privacy regulations and high annotation costs arising from domain expertise. Thus, existing benchmarks tend to underrepresent…
Shujun Xia, Haokun Lin, Yichen Wu, Yinan Zhou +4
Large Language Models (LLMs) hold great promise for healthcare applications, but fast-changing medical knowledge can quickly make their outputs outdated or inaccurate, limiting…
Wenjie Li, Yujie Zhang, Haoran Sun, Xingqi He +4
Long-form clinical videos are central to visual evidence-based decision-making, with growing importance for applications such as surgical robotics and related settings. However,…
Aofei Chang, Le Huang, Alex James Boyd, Parminder Bhatia +3
Medical large vision-language models (Med-LVLMs) have recently achieved remarkable progress in vision–language comprehension and medical image segmentation. However, existing…
Qiang Zhou, Hanzhen Zhu, Pan Wang, Rui Tu +4
According to the reformulated Learned Helplessness theory, repeated exposure to uncontrollable negative events can foster a depressogenic attributional style—increasing…
Boqiang Xu, Wei Zhang, Ding Ma, Jian Liang +2
Operating room (OR) scene graph generation (SGG) enables holistic modeling of OR domains by encoding interactions among medical staff, tools, and equipment as triplet-based…
Nassim Oufattole, Matthew B.A. McDermott, Collin Stultz
Autoregressive generative models for irregularly sampled clinical time-series data are increasingly used for zero-shot risk forecasting. Prior work typically adopts a single…
Itai Zilberstein, Ioannis Anagnostides, Zachary W. Sollie, Arman Kilic +1
Online matching has been a mainstay in domains such as Internet advertising and organ allocation, but practical algorithms often lack strong theoretical guarantees. We take an…
Marta Hasny, Laura Alexandra Daza, Keno Bressem, Maxime Di Folco +1
Large-scale medical biobanks provide imaging data complemented by extensive tabular information, such as clinical measurements or demographics. However, this abundance of tabular…
Shaohao Rui, Kaitao Chen, Weijie Ma, Xiaosong Wang
Extended Chain-of-Thought (CoT) reasoning has significantly bolstered the capabilities of medical large language models (LLMs). However, current models exhibit static…
Jiangtao Yan, Yanlin Qu, Yansheng Qiu, Shujian Gao +4
The longitudinal management of blinding fundus diseases constitutes a Partially Observable Markov Decision Process (POMDP) necessitating a critical precision-risk trade-off…
Zitao Shuai, Zongzhe Xu, David Yang, Wei Wang +1
Polysomnography (PSG) provides the gold standard for sleep assessment but suffers from substantial heterogeneity across recording devices and cohorts. There have been growing…
Zhengqiu Yu, Yueping Ding, Xiangrong Liu
Patient-level sepsis prediction requires models that track clinical deterioration over time and integrate heterogeneous structured evidence from electronic health records. We…
Maria Emilia Russo, Federico Di Valerio, Alessia Borghini, Alessio Ragno +1
Computational approaches have become central to Protein–Protein Interaction (PPI) research, complementing experimental techniques that remain costly and incomplete. While modern…
Jiwoong Sohn, Tomasz Sternal, Kenneth Styppa, Torsten Hoefler +1
Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require…
Jonathan Dan, Amirhossein Shahbazinia, Christodoulos Kechris, David Atienza
Reliable automatic seizure detection from long-term electroencephalogram recordings (EEG) remains an unsolved challenge, as current models often fail to generalize across patients…
Jinhan Liu, Mahsa Shoaran
Affective and cognitive disorders manifest as distributed, time-varying brain network dynamics across regions, channels, and time, challenging robust representation learning from…
Tianwei Lin, Zhongwei Qiu, Jie Cao, Jiang Liu +4
Medical vision-language models (VLMs) have rapidly advanced as general-purpose multimodal assistants, yet their deployment in 3D Computed Tomography (CT) analysis remains…
Marius Knorr, Robert Müller, Jan Peter Bremer, Nils Schweingruber
Fast Healthcare Interoperability Resources (FHIR) is the dominant standard for interoperable exchange of healthcare data. In FHIR, electronic health records form a directed graph…
Wentao Gao, Jiuyong Li, Lin Liu, Thuc Duy Le +3
Regional climate prediction presents unique challenges for time series foundation models, which typically process temporal patterns through single-pass inference. Expert…
Shuohao Gao, Xuanzhong Chen, Lingxiao Luo, Zilin Ding +3
Diagnosing complex diseases is inherently a sequential and iterative medical investigation process, in which a clinician strategically requests multiple rounds of diagnostic tests…
Jian Chen, Yipeng Du, Wenhao Yuan, Shuai Wang +4
Electrocardiogram (ECG) representation learning via ECG-report alignment is often hindered by the inherent structural and statistical divergence between signals and natural…
Yuqiao Meng, Luoxi Tang, Dazheng Zhang, Rafael Brens +4
The rapid adoption of large language models (LLMs) in digital health has been driven by a "scaling-first" philosophy, i.e., the assumption that clinical intelligence increases…
Zhenglun Kong, Mufan Qiu, John Boesen, xiang lin +4
Understanding how cellular morphology, gene expression, and spatial context jointly shape tissue function is a central challenge in biology. Image-based spatial transcriptomics…
Ying Wang, Guoheng Huang, Chan-Tong Lam, Xiaochen Yuan
Reliable medical audio diagnosis requires models that are both accurate and honest about their uncertainty. However, fine-tuned models on small, imbalanced datasets often become…
Yuanlin Yang, Chenhui Li, Xuhao Guo, Anqi Zhang +2
Biomedical regression tasks require predicting continuous targets from heterogeneous and unstructured evidence. While Large Language Models (LLMs) provide a robust interface for…
Ana Sanchez-Fernandez, Thomas Pinetz, Werner Zellinger, Günter Klambauer
Biomedical imaging data presents enormous potential for deep learning models to predict invaluable properties, such as diseases and drug effects. However, unavoidable alterations…
Hongyu Shi, Kaizhong Zheng, Wensheng Zhai, Shuai Jiang +2
Multi-modal neuroimaging-based psychiatric diagnosis must integrate cross-modal agreement with modality-specific complementarity, yet in real multi-site cohorts these signals are…
Maxx Richard Rahman, Mostafa Hammouda, Wolfgang Maass
Large Language Models have shown strong generalization across natural language tasks but remain underexplored for longitudinal clinical profiles. In sports anti-doping, biological…
Weiren zhao, DONG Yi, Cheng Chen
Unifying multimodal understanding and generation is a compelling frontier that is beginning to emerge in the medical field. However, the limited existing unified medical models…
Bong Gyun Kang, Junyong Ahn, Hyeongrok Han, Sungroh Yoon
Electronic Health Records (EHRs) possess unique characteristics distinct from natural language, yet existing EHR foundation models often rely on suboptimal NLP-based approaches.…
Kaitao Chen, Weiqian Zhao, Jiamin Wu, Qihao Zheng +4
Vision-language models (VLMs) combining reinforcement learning (RL) ignite remarkable progress in multimodal reasoning, yet still struggle with medical images, which typically…
Shohaib Shaffiey, Massimiliano Pierobon
The fields of AI-based disease fingerprinting, drug discovery and repurposing are currently among the emerging frontiers of machine learning applied to medicine. One major…
Junzhi Ning, Wei Li, Cheng Tang, Jiashi Lin +4
Medical workflows routinely combine reading images with producing visual and textual outputs, making both image understanding and generation central to medical AI. Most existing…
Chenyang Xu, Dezhen Wang, Hao Wang
Synthesizing authentic phonocardiograms (PCG) from ubiquitous electrocardiograms (ECG) is a critical task for accessible cardiac monitoring. Existing generative models, however,…
Metadata from BioTender-max/icml2026-ai-bio (CC0-1.0).