Sheng Liu, PhD
I am a Postdoctoral Researcher at Stanford University, working with Prof. James Zou and Prof. Lei Xing. I earned my PhD in Data Science from New York University.
My research focuses on improving the robustness and reliability of AI models, agents, and systems, with an emphasis on robust pre-training and inference-time steering. I am particularly interested in their medical applications for dementia and oncology.
Outside of academia, I love playing tennis and am also a certified scuba diver and surfer.
Research Interests:
Latest News
2025
- 2025 02Co-authored OctoTools: An Agentic Framework with Extensible Tools is online now. Demo and Github ⭐️
- 2025 02TextGrad is now accepted by Nature .
- 2025 02Sheng et al. Reducing hallucinations in VLM via latent space steering is accepted at ICLR 2025 as a Spotlight paper. (Code)
- 2025 02Paper on robust federated learning Data Reconstruction Attacks and Defenses is accepted at AISTATS 2025.
- 2025 02Large scale multimodal dataset for medicine MedTrinity-25M is accepted at ICLR 2025.(Code, Data)
- 2025 01I am co-organizing the "Multimodal Foundation Models for Biomedicine" workshop at ICCV 2025.
- 2025 01Guesture lecture at UC Santa Cruz by Prof. Yuyin Zhou
2024
- 2024 09I am co-organizing the GenAI for Healthcare workshop at NeurIPS 2024.
- 2024 09One paper Training-Free Guidance for Diffusion Models is accepted at NeurIPS 2024 as spotlight.
- 2024 07 LLMs in scientific papers is accepted at COLM 2024.
- 2024 07Our paper on AI guided radiotherapy treatment planning is online Arxiv.
- 2024 06We introduce TextGrad: Automatic "Differentiation" via Text! Start optimizing prompts in your LLM system.Fetching stars...
- 2024 06 In-Context Vector: making ICL more effective and controllable has been accepted at ICML 2024.
- 2024 06 Impact of ChatGPT in AI review has been accepted at ICML 2024 (oral).
- 2024 03New York Times opinion on AI-generated articles (co-authored manuscript).
- 2024 03Two papers Theoretical analysis on neural collapse and Robust self-supervised learning are accepted at TMLR.
Active Research Areas
1. Robust and Reliable AI
The real world is complex and noisy: data can be imperfect, labels may be inaccurate, and user prompts are often ambiguous. Directly deploying AI models in such environments without careful consideration can be dangerous. We develop methods to enhance the robustness and reliability of AI models by improving supervised learning with noisy labels and advancing the safety and alignment of foundation models for real-world applications.
Related Publications:
Reducing hallucinations in vision-language models via latent space steering
Sheng Liu, Haotian Ye, Lei Xing, James Zou
ICLR (2025) Spotlight
In-context vector: making in-context learning more effective and controllable
Sheng Liu, Haotian Ye, Lei Xing, James Zou
ICML (2024)
Early-learning regularization prevents memorization of noisy labels
Sheng Liu, Jonathan Niles-Weed, Narges Razavian, Carlos Fernandez-Granda
NeurIPS (2020)
2. AI Systems and AI agents
We develop AI software platforms to assist human experts in clinical practice and promote human–AI collaboration. We also optimize AI systemts with large language models (LLMs).
Related Publication:
Optimizing generative AI by backpropagating language model feedback
Mert Yuksekgonul*, Federico Bianchi*, Joseph Boen*, Sheng Liu*, Pan Lu*, Zhi Huang*, Carlos Guestrin, James Zou (*equal contribution)
Nature (2025)
OctoTools: An Agentic Framework with Extensible Tools for Complex Reasoning
Pan Lu* , Bowen Chen*, Sheng Liu*, Rahul Thapa, Joseph Boen, James Zou (*equal contribution)
Arxiv (2025)
3. AI for medicine
Medicine presents high-stakes, complex challenges where accuracy and reliability are paramount. We develop AI models to support clinical decision-making in areas such as radiation oncology and Alzheimer’s disease, aiming to improve treatment planning, diagnosis, and patient outcomes. Our work integrates domain knowledge with advanced machine learning techniques to ensure safe, effective, and trustworthy AI solutions for real-world medical applications.
Related Publication:
Automated radiotherapy treatment planning guided by GPT-4Vision
Sheng Liu*, Oscar Pastor-Serrano*, Yizheng Chen, Matthew Gopaulchan, Weixing Liang, Mark Buyyounouski, Erqi Pollom, Quynh-Thu Le, Michael Gensheimer, Peng Dong, Yong Yang, James Zou, Lei Xing (*equal contribution)
Arxiv (2024)
Generalizable deep learning model for early Alzheimer’s disease detection from structural MRIs
Sheng Liu, Arjun V Masurkar, Henry Rusinek, Jingyun Chen, Ben Zhang, Weicheng Zhu, Carlos Fernandez-Granda, Narges Razavian
Nature Scientific Reports (2023)