Zhi Huang, PhD

Hi! I am a tenure-track assistant professor at the University of Pennsylvania started in 2025, with multiple appointments & affiliations in the Perelman School of Medicine. I received my PhD in Electrical and Computer Engineering from Purdue University in August 2021. From August 2021 to June 2024, I was a postdoctoral fellow at Stanford University, working with Prof. James Zou and Prof. Thomas Montine.

My research focuses on AI/ML innovation and its application to medicine, with topics including vision-language foundation model for pathology (Nature Medicine'23 cover article), human-AI collaboration (Nature BME'24 cover article), neurodegenerative diseases (Nature Communications'23), optimizing LLMs (Nature'25), etc. My research has drawn wide public attention (including the New York Times, Stanford Magazine, and Stanford Scope) and has resulted in translational innovations. In 2022, my postdoc mentors and I co-founded nuclei.io — a human-in-the-loop AI platform for digital pathology.

I am very fortunate to work with many amazing students and researchers. We are always looking for talented folks to join us. Feel free to contact me or visit Openings if you are interested to learn more about our research!

Affiliations:

Assistant Professor

Perelman School of Medicine

Pathology & Laboratory Medicine (primary)

Department of Biostatistics, Epidemiology and Informatics (secondary)

Faculty Affiliation

AI2D center

Institute for Translational Medicine and Therapeutics (ITMAT)

Penn Institute for Biomedical Informatics

Graduate Group Affiliation

Electrical and Systems Engineering (ESE)

BioEngineering (BE)

Research Interests:

AI for healthcare
Human-AI collaboration
Computational pathology
Precision medicine

Latest News

2025

202504 nuclei.io is featured on Nature Biomedical Engineering April 2025 cover story & Nature Outlook on Medical Diagnostics.
202503 Our LLM optimization work is published in Nature.
202501 I am co-organizing the "Multimodal Foundation Models for Biomedicine" workshop at CVPR 2025. See you in Nashville!
202501 LLM feedback for federated learning is accepted at ICLR 2025.

2024

202411 I am serving as a program committee member for RECOMB 2025.
202407 LLMs in scientific papers is accepted at COLM 2024. See you in Philadelphia!
202406 Huang et al. Pathologist-AI collaboration study is published in Nature Biomedical Engineering.
202406 We introduce TextGrad: Automatic "Differentiation" via Text! Start optimizing prompts in your LLM system.
202405 Impact of ChatGPT in AI review is accepted at ICML 2025 (oral).
202403 New York Times opinion on AI-generated articles (co-authored manuscript).
202401 New study on off-label and off-guideline cancer therapy usage is accepted in Cell Reports Medicine.

2023

202311 Flash talk at Stanford Pathology (Video).
202310 Huang et al. invited commentary on resilience to AD is published in Neuroscience Insights.
202309 Visual-language AI for pathology is featured on Nature Medicine September 2023 cover story.
202308 Huang, Bianchi et al. Visual-language AI for pathology is published in Nature Medicine.
202305 Huang et al. Brain proteomic analysis is published in Nature Communications.
202303 nuclei.io is spotlighted in the Stanford Catalyst 2023 cohort [News].
202301 Huang et al. Multi-modal pathology imaging analysis is published in NPJ Precision Oncology.

Active Research Areas

1. Foundation Model for Pathology

Pathology is medicine's ground truth. We train AI models with vision, language, and knowledge to improve machine understanding of pathology.

Related Publication:

A visual–language foundation model for pathology image analysis using medical Twitter

Zhi Huang†, Federico Bianchi†, Mert Yuksekgonul, Thomas J. Montine, James Zou (†: Equal contribution)

Nature Medicine (2023), cover article

✨ Checkout our OpenPath, a visual-language dataset for pathology at here.

[paper] [website] [blog]

2. AI Systems and Human-AI Collaboration

We develop AI software platforms to assist human experts in clinical practice, promote human-AI collaboration. We also optimize LLMs.

Related Publication:

A pathologist–AI collaboration framework for enhancing diagnostic accuracies and efficiencies

Zhi Huang, Eric Yang, Jeanne Shen, Dita Gratzinger, ..., Thomas J. Montine & James Zou

Nature Biomedical Engineering (2024), cover article

nuclei.io: AI platform for digital pathology [website] []

[research website] [paper] [news]

Optimizing generative AI by backpropagating language model feedback

Mert Yuksekgonul*, Federico Bianchi*, Joseph Boen*, Sheng Liu*, Pan Lu*, Zhi Huang*, Carlos Guestrin, James Zou

Nature (2025)

✨ We introduce TextGrad, automatic "differentiation" via text.

[paper] [

]

Read the news: [Stanford HAI] [MarkTechPost] [Synced]

3. Machine Learning and Statistical Analysis for Human Diseases

We leverage machine learning and statistical analysis tools to study human diseases (e.g., Alzheimer's Disease, breast cancer) with imaging, omics, and clinical data.

Related Publication:

Brain Proteomic Analysis Implicates Actin Filament Processes and Injury Response in Resilience to Alzheimer's Disease

Zhi Huang, Gennifer E. Merrihew, Eric B. Larson, Jea Park,..., James Y. Zou, Michael J. MacCoss & Thomas J. Montine

Nature Communications (2023)

✨ First study to identify molecular signatures of resilience to Alzheimer's Disease.

[paper]

Artificial Intelligence Reveals Features Associated with Breast Cancer Neoadjuvant Chemotherapy Responses from Multi-stain Histopathologic Images

Zhi Huang, et al.

NPJ Precision Oncology (2023)

✨ AI-powered prediction of breast cancer treatment outcomes from H&E and IHC.

[paper] [news]

SALMON: Survival Analysis Learning with Multi-Omics Neural Networks on Breast Cancer

Zhi Huang, et al.

Frontiers in genetics (2019)

✨ Novel deep learning framework for multi-omics data integration and survival prediction.

[paper] [

]