James Zou
Associate Professor of Biomedical Data Science
Develops machine learning methods for biomedicine and studies the reliability of AI models.
James Zou's lab at Stanford works on trustworthy & safe ML, medical image analysis and data management & big data. 131 papers since 2023, 74 at top venues, mostly NeurIPS and ICML. 17 PhD students. 25 alumni, most often to AI startup and OpenAI. Applications close .
At a glance
- papers, 2023–2026
- 34/ year
- at T1 venues
- 56%
- PhD students
- 17
- MS students
- 0
- alumni listed
- 25
Papers by yearfrom the lab homepage, 2023–2026
- 1 papers2026 · T1 0 · collecting
- 36 papers2025 · T1 27
- 53 papers2024 · T1 30
- 41 papers2023 · T1 17
T1 venuesOtherVenues: NeurIPS, ICML, ICLR, Nature, Nature Communications
Research focuspaper titles that use each topic's words
- Medical image analysis22 papers
- Data management & big data20 papers
- Trustworthy & safe ML1 paper
- biomedical data0 papers
- reliability0 papers
Paper list
From the lab homepage.
20261 paper
202536 papers
- NeurIPS (T1)4KAgent: Agentic Any Image to 4K Super-Resolution. (opens in a new tab)
- NeurIPS (T1)A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learning. (opens in a new tab)
- CirculationADMET-AI Enables Interpretable Predictions of Drug-Induced Cardiotoxicity. (opens in a new tab)
- Nature Communications (T1)An automated framework for assessing how well LLMs cite relevant medical references. (opens in a new tab)
- NeurIPS (T1)AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration. (opens in a new tab)
- International Conference on Learning Representations (ICLR). OralCapturing the Temporal Dependence of Training Data Influence. (opens in a new tab)
- NeurIPS (T1)CGBench: Benchmarking Language Model Scientific Reasoning for Clinical Genetics Research. (opens in a new tab)
- Nature Communications (T1)Characterizing mutation-treatment effects using clinico-genomics data of 78,287 patients with 20 types of cancers. (opens in a new tab)
- ICML (T1)CollabLLM: From Passive Responders to Active Collaborators. (opens in a new tab)
- NeurIPS (T1)EvoLM: In Search of Lost Language Model Training Dynamics. (opens in a new tab)
- NeurIPS (T1)ExGra-Med: Extended Context Graph Alignment for Medical Vision-Language Models. (opens in a new tab)
- Nature BiotechnologyExploring the use of AI authors and reviewers at Agents4Science. (opens in a new tab)
202453 papers
- Nature MedicineA generalist vision-language foundation model for diverse biomedical tasks. (opens in a new tab)
- Nature Biomedical EngineeringA pathologist–AI collaboration framework for enhancing diagnostic accuracies and efficiencies. (opens in a new tab)
- NeurIPS (T1)Accelerating Transformers with Spectrum-Preserving Token Merging. (opens in a new tab)
- BioinformaticsADMET-AI: a machine learning ADMET platform for evaluation of large-scale chemical libraries. (opens in a new tab)
- NeurIPS (T1)Are More LLM Calls All You Need? Towards the Scaling Properties of Compound AI Systems. (opens in a new tab)
- ICML (T1)ArtWhisperer: A Dataset for Characterizing Human-AI Interactions in Artistic Creations. (opens in a new tab)
- ICML (T1)Assessing the Impact of ChatGPT in AI Conference Peer Reviews. (opens in a new tab)
- NeurIPS (T1)AvaTaR: Optimizing LLM Agents for Tool-Assisted Knowledge Retrieval. (opens in a new tab)
- npj Digital MedicineBridging the literacy gap for surgical consents: an AI-human expert collaborative approach. (opens in a new tab)
- NEJM AICan Large Language Models Provide Useful Feedback on Research Papers? A Large-Scale Empirical Analysis. (opens in a new tab)
- NeurIPS Datasets and BenchmarksCARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language Models. (opens in a new tab)
- Nature (T1)ChatGPT is transforming peer review — how can we use it responsibly? (opens in a new tab)
202341 papers
- PNAS Nexus7-UP: generating in silico CODEX from a small set of immunofluorescence markers. (opens in a new tab)
- Medical Image Analysis (T1)A clinically applicable AI system for diagnosis of congenital heart diseases based on computed tomography images. (opens in a new tab)
- npj Digital MedicineA deep learning-based electrocardiogram risk score for long term cardiovascular death and disease. (opens in a new tab)
- Nature Communications. JSM Outstanding Paper AwardA spectral method for assessing and combining multiple data visualizations. (opens in a new tab)
- Nature Medicine. Cover articleA visual–language foundation model for pathology image analysis using medical Twitter. (opens in a new tab)
- Journal of Emergency MedicineAI-enabled assessment of cardiac function and video quality in emergency department point-of-care echocardiograms. (opens in a new tab)
- npj Precision OncologyAnalyses of canine cancer mutations and treatment outcomes using real-world clinico-genomics data of 2119 dogs. (opens in a new tab)
- NeurIPS (T1)Beyond Confidence: Reliable Models Should Also Consider Atypicality. (opens in a new tab)
- Findings of ACLBeyond Positive Scaling: How Negation Impacts Scaling Trends of Language Models. (opens in a new tab)
- Nature (T1)Blinded, randomized trial of sonographer versus AI cardiac function assessment. (opens in a new tab)
- Nature Communications (T1)Brain proteomic analysis implicates actin filament processes and injury response in resilience to Alzheimer’s disease. (opens in a new tab)
- New England Journal of Medicine AICharacterizing the Clinical Adoption of Medical AI Devices through U.S. Insurance Claims. (opens in a new tab)
Where alumni went25 listed, 71% to industry
Industry 12Academia 5
Most often to
- AI startup
- OpenAI
- AI for cardiology
Also to
- Carnegie Mellon University
- MIT
- Stanford University
Apply to StanfordFall 2027
Stanford Computer Science PhD
- Round
- Single round
- Opens
- GRE
- See program page
- English
- TOEFL/IELTS
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- SGNUS
Bryan Hooi
- Time series, anomaly & spatio-temporal data
- GBImperial
Alessio Lomuscio
- Reasoning & neuro-symbolic AI
- GBImperial
Francesco Belardinelli
- Reasoning & neuro-symbolic AI
- USUT Austin
Qiang Liu
- Learning theory & optimization
- CNSJTU
Lei Xu
- Learning theory & optimization
- CNTsinghua
Guiguang Ding
- Efficient training & inference
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Stanfordcloses