Zifeng Wang
I am currently pursuing a PhD in Computer Science at Sunlab@UIUC, under the guidance of Prof. Jimeng Sun.
My research primarily revolves around the development of foundation models, including large language models (LLMs), tabular models, vision-language models, and biomedical multimodal models, for drug discovery and development.
My recent focus has shifted towards creating LLM-driven systems to enhance human-AI collaboration in various clinical trial tasks, including trial search, trial design, and data analysis.
We are actively developing a startup venture, Keiji.AI, with the vision to revolutionize clinical trial development with cutting-edge AI techniques.
News
Education
Experience
Activity & Award
Service
PC Member/Reviewer for NeurIPS'[22,23,24], EMNLP'[22,23], IJCAI'[22,23,24], AAAI'[22,24,25], NAACL'24, COLM'24, ICLR'24, ACL'23, KDD'23, NLPCC'24, ICIP'21, RECOMB'25, ICASSP'21.
Reviewer for TPAMI, JAIR, JAMIA, Bioinformatics, ACM Computing Surveys.
Teaching Experience
Invited Talk
- Aug, 2025, Invited talk: "Large Language Models for Clinical Trial Search, Recruitment, and Design", in Invited Program@Takeda in JSM'25.
Mar, 2024, Bridging Biomedical Foundation Models, invited by Xuegong Lab@Tsinghua University.
Oct, 2023, Bridging Biomedical Foundation Models, invited by AI4Science@ByteDance.
Oct, 2023, Automate clinical trial design with large language models, invited by Medidata.
Aug, 2023, Medical Vision-Language Modeling, invited by Prof. Lee@UToronto.
July, 2023, Medical Vision-Language Modeling, invited by Stanford MedAI Group.
July, 2023, Unifying Large Language Models and Knowledge Graphs, invited by Amazon Titan Lab.
Apr, 2023, Medical Vision-Language Modeling from Unpaired Images and Texts, invited by National Center for Biotechnology Information (NCBI-NLM-NIH).
Feb, 2023, Advances in Transfer Learning for Tabular Prediction (CN), invited by AI Time.
Nov, 2022, How to do transfer learning and zero-shot learning on the tabular domain? (CN), invited by AI Time
April, 2022, Understanding Deep Learning via Information in Weights, invited by AI Time.
March, 2022, PAC-Bayes Information Bottleneck, invited by ReadPaper.
|