Final-year PhD at MMLab@NTU, advised by Prof. Ziwei Liu.
I build and evaluate multimodal large language models, with a focus on post-training and benchmarking.
New Video-TT ICCV'25 · LLaVA-Video & OneVision TMLR'25
Current Interests
Information-dense Video Reasoning: Building models that identify salient details, connect dispersed evidence, and infer causal structure in complex visual narratives.
Harness-centric LLM Systems: Designing harnesses that structure memory for effective in-context comprehension and encode success criteria as executable checks that test for correctness, not mere plausibility.
Benchmarking Model–Human Gaps: Developing diagnostic benchmarks that reveal systematic differences between models and humans, and suggest better optimization targets.
Yuanhan Zhang*, Yunice Chew*, Yuhao Dong, Aria Leo, Bo Hu, Ziwei Liu
A video reasoning benchmark that eliminates key-frame sampling bias — exposes a 48-point gap between GPT-4o (36.6%) and humans (84.3%).
Yuanhan Zhang, Jinming Wu, Wei Li, Bo Li, Zejun Ma, Ziwei Liu, Chunyuan Li
An open video LMM built entirely on synthetic data — 1.2M+ HuggingFace downloads.
Bo Li, Yuanhan Zhang, Dong Guo, Renrui Zhang, Feng Li, Hao Zhang, Kaichen Zhang, Yanwei Li, Ziwei Liu, Chunyuan Li
One model for image, multi-image, and video — SOTA across all three.
Yuan Liu*, Haodong Duan*, Yuanhan Zhang*, Bo Li*, Songyang Zhang*, Wangbo Zhao, Yike Yuan, Jiaqi Wang, Conghui He, Ziwei Liu, Kai Chen, Dahua Lin
The go-to multimodal evaluation benchmark — adopted by Qwen-VL, InternVL, LLaVA, and more.
