Hejie Cui

Applied Scientist, Amazon SFAI Rufus LLM & Postdoctoral Researcher

Amazon Store Foundation AI (SFAI)

Stanford University

Biography

Hello! I am currently an Applied Scientist at Amazon Store Foundational AI (SFAI), where I work on large-scale LLM post-training and alignment for Amazon’s Rufus LLM. My work focuses on reinforcement learning, instruction fine-tuning, synthetic data generation, and evaluation methods to improve LLM reasoning, reliability, and controllability.

Previously, I was a Postdoctoral Researcher at Stanford University, advised by Prof. Nigam H. Shah and Prof. Sanmi Koyejo, where I worked on LLM post-training, synthetic data pipelines, and benchmark development for health applications.

I received my Ph.D. in Computer Science from Emory University, advised by Prof. Carl Yang and working closely with Prof. Joyce C Ho. I obtained my bachelor’s degree (Valedictorian, GPA Ranking: 1/164) in Software Engineering at Tongji University, where I also conducted research in computer vision and machine learning.

News

[2025.12] Our work MedHELM: Holistic Evaluation of Large Language Models for Medical Tasks has been officially accepted for publication in Nature Medicine. Huge thanks to all collaborators for making this possible!
[2025.08] Our work on temporal instruction tuning (𝗧𝗜𝗠𝗘𝗥⌛️) has been officially accepted for publication in npj Digital Medicine. Grateful to Dr. Shah and all collaborators for their valuable contributions and support!
[2025.07] Our work CuraBench: A Benchmark Dataset Generation System for Healthcare AI Evaluation is accepted to KDD 2025 HealthDay. Thanks to all the collaborators!
[2025.06] One paper on knowledge-enhanced reasoning is accepted to AMIA 2025.
[2025.06] Our survey on knowledge graphs for healthcare is accepted to Journal of Biomedical Informatics. Congrats!
[2025.06] Our work on multimodal EHR is accepted to MedInfo 2025.
[2025.05] Our work 𝗠𝗲𝗱𝗛𝗘𝗟𝗠: Holistic Evaluation of Large Language Models for Medical Tasks is now available online. Thank you to all the coauthors for their hard work!
[2025.05] Our work CLIMB: Data Foundations for Large Scale Multimodal Clinical Foundation Models is accepted to ICML 2025. Congrats Wei!
[2025.05] Our work on Domain Specialization as the Key to Make Large Language Models Disruptive: A Comprehensive Survey is accepted to ACM Computing Surveys.
[2025.03] We build 𝗠𝗲𝗱𝗛𝗘𝗟𝗠✨: a comprehensive benchmark evaluating AI on realistic clinical tasks that healthcare professionals perform daily instead of just medical exams. Check out our HAI blogpost and MedHELM leaderboard for more details.
[2025.02] We release three de-identified, longitudinal EHR datasets from Stanford: more details — now freely available for non-commercial research-use worldwide.
[2025.02] Pleased to share that our grant proposal 𝗖𝘂𝗿𝗮𝗕𝗲𝗻𝗰𝗵 has been selected for funding through by the Stanford RAISE Health Seed Grant Program and Stanford HAI. A big thank you to all the collaborators for their support!
[2024.12] Served as a Junior Chair at the Foundations Models and Multimodal AI Round Table at ML4H 2024 Symposium. The topic summary can be found here.
[2024.09] Our work on clinician preference aligned synthetic instruction generation for visual instruction tuning is accepted to NeurIPS'24 Datasets and Benchmarks Track.
[2024.09] Our work on network recall by large language models is accepted to NeurIPS'24 Research Track. Abstract version is presented on IC2S2'24 as an Oral.
[2024.08] Excited to be selected as a Rising Star Spotlight Speaker to give a talk about my research at University of Michigan 2024 AI Symposium.
[2024.06] Our work on multi-agent LLM reasoning for EHR-based few-shot disease prediction is accepted to AMIA Annual Symposium as an Oral.
[2024.05] Our work on disease subtyping is accepted to KDD Applied Data Science Track.
[2024.04] My PhD Work is selected to the CHIL Doctoral Symposium. Thank you CHIL!
[2024.04] I successfully defended my dissertation. Officially, Dr. Cui!
[2024.03] Our survey paper on LLM domain specialization is cited by the 2024 Economic Report of the President!
[2023.12] Glad to received NSF Student Travel Support Award for the ICDM 2023.
[2023.09] Our paper on artificial node features on non-attributed graphs has been selected as Most Influential CIKM Papers produced by Best Paper Digest.
[2023.09] One work on visual knowledge extraction is accepted to NeurIPS'23.
[2023.09] Two work are accepted to PSB'24. Congrats to Alexis (High-Schooler)!
[2023.08] Humbled to be selected as a 2023 EECS Rising Star!
[2023.05] One work on biological data augmentation is accepted to KDD'23.
[2023.05] One work on multimodal extraction is accepted to ACL'23 Findings.
[2023.04] Our work on brain network pre-training is accepted to CHIL'23 as an Oral.
[2022.11] Our paper on few-shot learning is accepted to AAAI'23 as an Oral.
[2022.11] Glad to receive NeurIPS AI4Science Travel Award!
[2022.10] Our Benchmark paper on Graph Neural Networks for brain networks has now been officially accepted to IEEE TMI.
[2022.09] Our paper on brain transformer is accepted to NeurIPS'22 as an Spotlight.
[2022.08] One paper on node feature for non-attributed graphs is accepted to CIKM'22.
[2022.06] One paper on interpretable GNNs is accepted to MICCAI'22 as an Oral.

Interests

LLM Post-training and Alignment (RL/IFT)
Multimodal LLMs
LLM Evaluation and Benchmarking
AI for Healthcare

Education

Ph.D. in Computer Science, 2019-2024
Emory University
B.Eng. in Software Engineering, 2015-2019
Tongji University

Selected Publications

Suhana Bedi†, Hejie Cui†, Miguel Fuentes†, Alyssa Unell†, Michael Wornow, Juan M. Banda, Nikesh Kotecha, Timothy Keyes, Yifan Mai, Mert Oez, Hao Qiu, Shrey Jain, Leonardo Schettini, Mehr Kashyap, Jason Alan Fries, Akshay Swaminathan, Philip Chung, Fateme Nateghi Haredasht, Ivan Lopez, Asad Aali, Gabriel Tse, Ashwin Nayak, Shivam Vedak, Sneha S. Jain, Birju Patel, Oluseyi Fayanju, Shreya Shah, Ethan Goh, Dong-han Yao, Brian Soetikno, Eduardo Reis, Sergios Gatidis, Vasu Divi, Robson Capasso, Rachna Saralkar, Chia-Chun Chiang, Jenelle Jindal, Tho Pham, Faraz Ghoddusi, Steven Lin, Albert S. Chiou, Christy Hong, Mohana Roy, Michael F. Gensheimer, Hinesh Patel, Kevin Schulman, Dev Dash, Danton Char, Lance Downing, Francois Grolleau, Kameron Black, Bethel Mieso, Aydin Zahedivash, Wen-wai Yim, Harshita Sharma, Tony Lee, Hannah Kirsch, Jennifer Lee, Nerissa Ambers, Carlene Lugtu, Aditya Sharma, Bilal Mawji, Alex Alekseyev, Vicky Zhou, Vikas Kakkar, Jarrod Helzer, Anurang Revri, Yair Bannett, Roxana Daneshjou, Jonathan Chen, Emily Alsentzer, Keith Morse, Nirmal Ravi, Nima Aghaeepour, Vanessa Kennedy, Akshay Chaudhari, Thomas Wang, Sanmi Koyejo, Matthew P. Lungren, Eric Horvitz, Percy Liang, Michael A. Pfeffer, Nigam H. Shah (2025). MedHELM: Holistic Evaluation of Large Language Models for Medical Tasks. Nature Medicine.

Code Project

Hejie Cui†, Alyssa Unell†, Bowen Chen, Jason Alan Fries, Emily Alsentzer, Sanmi Koyejo, Nigam H. Shah (2025). TIMER: Temporal Instruction Modeling and Evaluation for Longitudinal Clinical Records. npj Digital Medicine.

PDF

Hejie Cui, Alyssa Unell, Haoran Zhang, Caleb Winston, Jason Alan Fries, Sanmi Koyejo, Nigam H. Shah (2025). CuraBench: A Benchmark Dataset Generation System for Healthcare AI Evaluation. KDD 2025 HealthDay Blue Sky Ideas Track.

PDF

Hejie Cui†, Jiaying Lu†, Ran Xu†, Shiyu Wang, Wenjing Ma, Yue Yu, Shaojun Yu, Xuan Kan, Chen Ling, Liang Zhao, Zhaohui S. Qin, Joyce C. Ho, Tianfan Fu, Jing Ma, Mengdi Huai, Fei Wang, Carl Yang (2025). A Review on Knowledge Graphs for Healthcare: Resources, Applications, and Promises. Journal of Biomedical Informatics (JBI).

Wei Dai, Peilin Chen, Malinda Lu, Daniel Li, Haowen Wei, Hejie Cui, Paul Pu Liang (2025). CLIMB: Data Foundations for Large Scale Multimodal Clinical Foundation Models. The International Conference on Machine Learning (ICML).

PDF Code Dataset

Hejie Cui, Xinyu Fang, Ran Xu, Xuan Kan, Joyce C. Ho, Carl Yang (2025). Multimodal Fusion of EHR in Structures and Semantics: Integrating Clinical Records and Notes with Hypergraph and LLM. International Medical Informatics Conference (MedInfo).

Hejie Cui†, Lingjun Mao†, Xin Liang, Jieyu Zhang, Hui Ren, Quanzheng Li, Xiang Li, Carl Yang (2024). Biomedical Visual Instruction Tuning with Clinician Preference Alignment. The Conference on Neural Information Processing Systems (NeurIPS).

PDF Code Dataset Project Source Document

Yanbang Wang, Hejie Cui, Jon Kleinberg (2024). Microstructures and Accuracy of Graph Recall by Large Language Models. The Conference on Neural Information Processing Systems (NeurIPS). Abstract version accepted to International Conference on Computational Social Science (IC2S2) (Oral).

Hejie Cui, Zhuocheng Shen, Jieyu Zhang, Hui Shao, Lianhui Qin, Joyce C Ho, Carl Yang (2024). LLMs-based Few-Shot Disease Predictions using EHR: A Novel Approach Combining Predictive Agent Reasoning and Critical Agent Instruction. American Medical Informatics Association (AMIA) Annual Symposium (Oral).

Ziyang Zhang, Hejie Cui, Ran Xu, Yuzhang Xie, Joyce C Ho, Carl Yang (2024). TACCO: Task-guided Co-clustering of Clinical Concepts and Patient Visits for Disease Subtyping based on EHR Data. ACM International Conference on Knowledge Discovery and Data Mining (KDD).

Hejie Cui, Xuan Kan, Xiaoxiao Li, Ying Guo, Lifang He, Liang Zhan, Carl Yang (2024). Brain Network Analysis with Graph Neural Network. Tutorial at the IEEE International Symposium on Biomedical Imaging (ISBI).

Ran Xu, Hejie Cui, Yue Yu, Xuan Kan, Wenqi Shi, Yuchen Zhuang, Wei Jin, Joyce C Ho, Carl Yang (2023). Knowledge-Infused Prompting: Assessing and Advancing Clinical Text Data Generation with Large Language Models. Annual Meeting of the Association for Computational Linguistics (ACL-Findings).

Hejie Cui†, Xinyu Fang†, Zihan Zhang, Ran Xu, Xuan Kan, Xin Liu, Manling Li, Yangqiu Song, Carl Yang (2023). Open Visual Knowledge Extraction via Relation-Oriented Multimodality Model Prompting. Neural Information Processing Systems (NeurIPS).

Xuan Kan, Zimu Li, Hejie Cui, Yue Yu, Ran Xu, Shaojun Yu, Zilong Zhang, Ying Guo, Carl Yang (2023). R-Mixup: Riemannian Mixup for Biological Networks. ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD).

Hejie Cui, Rongmei Lin, Nasser Zalmout, Chenwei Zhang, Jingbo Shang, Carl Yang, Xian Li (2023). PV2TEA: Patching Visual Modality to Textual-Established Information Extraction. Annual Meeting of the Association for Computational Linguistics (ACL-Findings).

Yi Yang, Hejie Cui, Carl Yang (2023). PTGB: Pre-Train Graph Neural Networks for Brain Network Analysis. The Conference on Health, Inference, and Learning (CHIL) (Oral).

Hejie Cui, Xuan Kan, Xiaoxiao Li, Lifang He, Liang Zhan, Ying Guo, Carl Yang (2023). Brain Network Analysis with Graph Neural Network. Tutorial at the International Conference on Intelligent Biology and Medicine (ICIBM).

Ran Xu, Yue Yu, Hejie Cui, Xuan Kan, Yanqiao Zhu, Joyce C Ho, Chao Zhang, Carl Yang (2023). Neighborhood-regularized Self-Training for Learning with Few Labels. The AAAI International Conference on Artificial Intelligence (AAAI) (Oral).

Hejie Cui, Wei Dai, Yanqiao Zhu, Xuan Kan, Antonio Aodong Chen Gu, Joshua Lukemire, Liang Zhan, Lifang He, Ying Guo, Carl Yang (2022). BrainGB: A Benchmark for Brain Network Analysis with Graph Neural Networks. IEEE Transactions on Medical Imaging (TMI) (IF: 11.037).

Xuan Kan, Wei Dai, Hejie Cui, Zilong Zhang, Ying Guo, Carl Yang (2022). Brain Network Transformer. The Conference on Neural Information Processing Systems (NeurIPS) (Spotlight).

Hejie Cui, Zijie Lu, Pan Li, Carl Yang (2022). On Positional and Structural Node Features for Graph Neural Networks on Non-attributed Graphs. The International Conference on Information and Knowledge Management (CIKM) (Most Influential CIKM Paper of 2022).

Hejie Cui, Wei Dai, Yanqiao Zhu, Xiaoxiao Li, Lifang He, Carl Yang (2022). Interpretable GNNs for Connectome-Based Brain Disorder Analysis. The International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) (Oral).

Xuan Kan, Hejie Cui, Joshua Lukemire, Ying Guo, Carl Yang (2022). FBNetGen: Task-aware GNN-based fMRI Analysis via Functional Brain Network Generation. The International Conference on Medical Imaging with Deep Learning (MIDL) (Oral).

Hejie Cui, Jiaying Lu, Yao Ge, Carl Yang (2022). How Can Graph Neural Networks Help Document Retrieval: A Case Study on CORD19 with Concept Map Generation. The European Conference on Information Retrieval (ECIR).

Xuan Kan, Hejie Cui, Carl Yang (2021). Zero-Shot Scene Graph Relation Prediction through Commonsense Knowledge Integration. The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML).

Hejie Cui, Xinglong Liu, Ning Huang (2019). Pulmonary Vessel Segmentation based on Orthogonal Fused U-Net++ of Chest CT Images. The International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI).

Hejie Cui

Applied Scientist, Amazon SFAI Rufus LLM & Postdoctoral Researcher

Amazon Store Foundation AI (SFAI)

Stanford University

Biography

News

Interests

Education

Selected Publications

Experience

Research Intern

Microsoft Research

Applied Scientist Intern

Amazon

Mitacs Global Research Intern

Queens University

Microsoft University Summer Camp

Microsoft Research Asia

SAP Student Training and Rotation Program

SAP

Selected Accomplishments

Talks

Services