Zhaolong Su

Ph.D. Student in Systems Engineering, Cornell University

About Me

I am a first-year PhD student at Cornell University (started 2026), supervised by Prof. Yifan Peng, focusing on Unified Models & World Models & Video Gen & Pre/Post-Training Methods. Before that, I Graduated from Beijing University of Technology (B.Eng. in Artificial Intelligence).

During my undergraduate studies, I was fortunate to work with Bloomberg Distinguished Prof. Alan Yuille at Johns Hopkins University, Prof. Wentao Zhang at Peking University. I also worked at HKU promoting CT-free scoliosis treatment. Somehow there's always something that keeps me busy; see sections News, Publications, Service & Invited Talks, and Awards & Experience for empirical evidence.

News

2026.09 First-author paper "Medical World Models for Disease Dynamics: A Literature Review" released. Paper.

2026.09 🍺 Honored to receive the AMD University Program Award.

2026.08 🍺 FedUMM was featured in the NVIDIA Developer Technical Blog.

2026.08 🍺 Triple-block Diffusion RWKV accepted by EMNLP findings.

2026.05 🏆 FedUMM received the Best Student Paper Award at FL@FM-WWW'26.

2026.05 Triplet-Block Diffusion RWKV released on arXiv.

2026.02 🍺 First tutorial The Road to Convergence: Evolution of Unified Multimodal Models accepted by CVPR'26, big thanks to all collaborators, will participate as main speaker.

2026.02 🍺 UniGame accepted by CVPR'26. See you in Denver!

2026.02 First-author paper "Consistency Should Be the Priority for Unified Multimodal Models" released.

2026.01 🍺 First-author paper FedUMM accepted by FL@FM-WWW'26. Congratulations!

2025.11 Attending NeurIPS 2025 in San Diego.

2025.11 First-author paper UniGame released on arXiv (CVPR'26 scoring 554).

Selected Publications

Unified Models

UniGame: Turning a Unified Multimodal Model Into Its Own Adversary

Zhaolong Su, Wang Lu, Hao Chen, Sharon Li, Jindong Wang.

CVPR'26 (Scoring 554)

Paper Code Website

FedUMM: A General Framework for Federated Learning with Unified Multimodal Models

Zhaolong Su, Leheng Zhao, Xiaoying Wu, Ziyue Xu, Jindong Wang.

FL@FM-WWW'26 Best Student Paper 🏆

Paper Code TechBlog

Consistency Should Be the Priority for Unified Multimodal Models

Zhaolong Su, Yinyi Luo, Yiqiao Jin, Mengqi Zhang, Wenyue Hua, Srijan Kumar, Qingsong Wen, Jindong Wang.

ICML 2026 (Under-review)

Paper

Language Models

Triplet-Block Diffusion RWKV

Ke Lin, Yiyang Luo, Zhaolong Su, Yunya Song, Anyi Rao.

EMNLP findings

Paper Code

Service · Invited Talk

Service

Peer-Review: ICLR'27, ACL'26, CVPR'26, CVPR'26 Workshop Journey to the Awards: Generative AI for Movie-Grade Video Production, ICML'26

Invited Talks

The Road to Convergence: Evolution of Unified Multimodal Models Lead Speaker, CVPR'26 Tutorial

Awards & Experience

AMD University Program Award. NOTE: GPU hours for a year.
FL@FM-WWW'26 Best Student Paper Award 2026.
NVIDIA Academic Grant Program 2025.
Silver Medal, China English Debate National Championship 2022.
China National Encouragement Scholarship 2022.
Outstanding Scholarship, BJUT 2021.
Class President,2021-2025.
National Level II Athlete (China). NOTE: I did martial arts for six years and two years of sanda, but I lost it now.

Album

I love postmodernism and wasteland punk-style photography, and I wish to be a great photographer someday.

BJUT — Beijing University of Technology, my beloved alma mater