Hanwen Xu

Hi, I'm an incoming tenure-track Assistant Professor in the Departments of Bioengineering and Computer and Information Science at the University of Pennsylvania, starting in spring 2027. I am currently a Visiting Faculty Researcher at Google. I earned my PhD from the Paul G. Allen School of Computer Science & Engineering at the University of Washington, where I was advised by Prof. Sheng Wang. Before that, I received my BS and MS from the Department of Automation at Tsinghua University. I have also benefited greatly from the longstanding mentorship and support of Dr. Hoifung Poon.

My research develops AI for healthcare and medicine, with a particular focus on (1) gigascale foundation models for healthcare, (2) multimodal patient representation learning, and (3) longitudinal modeling of patient health trajectories. More recently, I have been building foundation-model-powered agentic systems for cancer care. My work has been published in leading venues, including Nature, Cell, and NEJM AI. It has received broad media coverage, including from Forbes, Yahoo, and MIT Technology Review. The resulting models have been downloaded more than six million times worldwide and demonstrated translational impact in real-world clinical settings.

My research philosophy: a golden loop between AI and medicine.

AI-0 → Medicine (short-term): Stress-test general-purpose AI against real-world challenges.
Medicine → AI: We reshape AI with insights from real-world medicine.
AI → Medicine (long-term): Transform AI’s newfound capabilities into medicine’s new possibilities.

Lab Openings

I'm recruiting two PhD students to join my lab in the application cycle of Fall 2027. Admissions will be through UPenn Bioengineering or UPenn Computer Science, depending on your background and research interests. If you would like to join us on this journey, please fill out this Google form to tell me a bit about yourself. We look forward to hearing from you!

If you are looking for a research opportunity working with me (e.g., undergraduate, master's or PhD students), please email me directly with your CV and a brief description of your interests.

🔥 News

[2026/08] Introduce our new Flash series of foundation models. GigaPath-Flash, 50x smaller and retaining comparable performance with its initial version. GigaTIME-Flash, using ViT-S with clear gains on tumor microenvironment analysis.
[2026/07] Our cardiac specific foundation model for heart sizing has been accepted by Nature Communications. Congratulations!
[2025/12] Thrilled to announce that our GigaTIME paper, a multimodal AI system generating virtual population for tumor microenvironment modeling, has been published in Cell!
[2025/07] Our Prov-GigaPath model has been downloaded for more than two million times since its release on Hugging Face . Congratulations!
[2025/07] Happy to announce that Pisces, a drug combination model that integrates eight modalities, has been published in Cell Genomics !
[2025/07] Thrilled to announce that our Prov-GigaPath model published in Nature was used in the first real-world, real-time clinical deployment of digital pathology foundation models.
[2025/04] Excited to see that our model has been available on Azure AI foundary !
[2024/12] We are excited to bring LLaVA-Rad, a clinically accessible small multimodal radiology model, to the community! Please check our paper in Nature Communications.
[2024/05] We are excited to announce that the initial release of the Prov-GigaPath model and its code is now available. The GigaPath paper has been published in Nature .
[2024/03] We are excited to announce that BiomedCLIP has been published in NEJM AI .

📄 Publications

A whole-slide foundation model for digital pathology from real-world data

Hanwen Xu*, Naoto Usuyama*, Jaspreet Bagga, Sheng Zhang, Rajesh Rao, Tristan Naumann, Cliff Wong, Zelalem Gero, Javier González, Yu Gu, Yanbo Xu, Mu Wei, Wenhui Wang, Shuming Ma, Furu Wei, Jianwei Yang, Chunyuan Li, Jianfeng Gao, Jaylen Rosemon, Tucker Bower, Soohee Lee, Roshanthi Weerasinghe, Bill J. Wright, Ari Robicsek, Brian Piening, Carlo Bifulco, Sheng Wang, Hoifung Poon (* equal contribution)

Nature 2024 2,000,000 downloads

Paper Code Model 50+ Media Coverage

Multimodal AI generates virtual population for tumor microenvironment modeling

Jeya Maria Jose Valanarasu*, Hanwen Xu*, Naoto Usuyama*, Chanwoo Kim*, Cliff Wong, Peniel Argaw, Racheli Ben Shimol, Angela Crabtree, Kevin Matlock, Alexandra Q. Bartlett, Jaspreet Bagga, Yu Gu, Sheng Zhang, Tristan Naumann, Bernard A. Fox, Bill Wright, Ari Robicsek, Brian Piening, Carlo Bifulco, Sheng Wang, Hoifung Poon (* equal contribution)

Cell 2025

Paper Code Model Highlighted by Microsoft CEO

A Cardiac-specific CT Foundation Model for Heart Transplantation

Hanwen Xu, Addie Woicik, Sanaz Asadian, Junbo Shen, Zhengyan Zhang, Ali Nabipoor, J. Peter Musi, Jeffrey Keenan, Maziar Khorsandi, Bassel Al-Alao, Ioannis Dimarakis, Hamid Chalian, Yiing Lin, Daniel Fishbein, Jay Pal, Sheng Wang, Shin Lin

Nature Communications Accepted 2026

Code Model

Pisces: A multi-modal data augmentation approach for drug combination synergy prediction

Hanwen Xu*, Jiacheng Lin*, Addie Woicik, Zixuan Liu, Jianzhu Ma, Sheng Zhang, Hoifung Poon, Liewei Wang, Sheng Wang (* equal contribution)

Cell Genomics 2025

Code Model

A clinically accessible small multimodal radiology model and evaluation metric for chest X-ray findings

Juan Manuel Zambrano Chaves*, Shih-Cheng Huang*, Yanbo Xu*, Hanwen Xu*, Naoto Usuyama, Sheng Zhang, Fei Wang, Yujia Xie, Mahmoud Khademi, Ziyi Yang, Hany Awadalla, Julia Gong, Houdong Hu, Jianwei Yang, Chunyuan Li, Jianfeng Gao, Yu Gu, Cliff Wong, Mu Wei, Tristan Naumann, Muhao Chen, Matthew P Lungren, Akshay Chaudhari, Serena Yeung Levy, Curtis P Langlotz, Sheng Wang, Hoifung Poon (* equal contribution)

Nature Communications 2025 11,000 downloads

Code Model

A Multimodal Biomedical Foundation Model Trained from Fifteen Million Image–Text Pairs

Sheng Zhang*, Yanbo Xu*, Naoto Usuyama*, Hanwen Xu*, Jaspreet Bagga, Robert Tinn, Sam Preston, Rajesh Rao, Mu Wei, Naveen Valluri, Cliff Wong, Andrea Tupini, Yu Wang, Matt Mazola, Swadheen Shukla, Lars Liden, Jianfeng Gao, Angela Crabtree, Brian Piening, Carlo Bifulco, Matthew P. Lungren, Tristan Naumann, Sheng Wang, Hoifung Poon (* equal contribution)

The New England Journal of Medicine, AI 2024 4,500,000 downloads

Code Model

Deep flanking sequence engineering for efficient promoter design using DeepSEED

Pengcheng Zhang*, Haochen Wang*, Hanwen Xu*, Lei Wei, Liyang Liu, Zhirui Hu, Xiaowo Wang (* equal contribution)

Nature Communications 2023 Editor highlights

Code Media Coverage (新浪科技)

GraphPrompt: Graph-Based Prompt Templates for Biomedical Synonym Prediction

Hanwen Xu*, Jiayou Zhang*, Zhirui Wang*, Shizhuo Zhang, Megh Bhalerao, Yucong Liu, Dawei Zhu, Sheng Wang (* equal contribution)

Proceedings of the AAAI Conference on Artificial Intelligence 2023

Code

Multilingual translation for zero-shot biomedical classification using BioTranslator

Hanwen Xu, Addie Woicik, Hoifung Poon, Russ Altman, Sheng Wang

Nature Communications 2023

Code Media Coverage

ProTranslator: zero-shot protein function prediction using textual description

Hanwen Xu, Sheng Wang

International Conference on Research in Computational Molecular Biology 2022

Hanwen Xu

Warning

Action required

Lab Openings

🔥 News

📄 Publications