Zeqi Xiao

I am a final-year Ph.D. candidate at MMLab@NTU, Nanyang Technological University, fortunately supervised by Prof. Xingang Pan. Prior to this, I had a wonderful internship experience at Shanghai AI Laboratory, working with Dr. Jiangmiao Pang and Dr. Tai Wang. I received my Bachelor's degree from Zhejiang University.

I am on the job market, seeking research opportunities in multimodal learning and embodied intelligence. I would be glad to connect with teams working on related problems.

Email / Google Scholar / Twitter / Github

Research

My research aims to build models that understand, simulate, and interact with the dynamic visual world. I explore how data construction can expand model capabilities, and how long-term memory and efficient modeling can support sustained interaction. I translate desired capabilities into data and task designs that make them learnable and measurable.

	Video4Spatial: Towards Visuospatial Intelligence with Context-Guided Video Generation Zeqi Xiao, Yiwei Zhao, Lingxiao Li, Yushi Lan, Ning Yu, Rahul Garg, Roshni Cooper, Mohammad H. Taghavi, Xingang Pan CVPR, 2026 Findings project page / arXiv
	Trainable Log-linear Sparse Attention for Efficient Diffusion Transformers Yifan Zhou, Zeqi Xiao, Tianyi Wei, Shuai Yang, Xingang Pan CVPR, 2026 (Highlight) arXiv / code
	WorldMem: Long-term Consistent World Simulation with Memory Zeqi Xiao, Yushi Lan, Yifan Zhou, Wenqi Ouyang, Shuai Yang, Yanhong Zeng, Xingang Pan NeurIPS, 2025 project page / arXiv / code / demo / bibtex
	TokensGen: Harnessing Condensed Tokens for Long Video Generation Wenqi Ouyang, Zeqi Xiao, Danni Yang, Yifan Zhou, Shuai Yang, Lei Yang, Jianlou Si, Xingang Pan ICCV, 2025 project page / arXiv / bibtex
	Alias-Free Latent Diffusion Models: Improving Fractional Shift Equivariance of Diffusion Latent Space Yifan Zhou, Zeqi Xiao, Wenqi Ouyang, Shuai Yang, Xingang Pan CVPR, 2025 (Oral) project page / arXiv / code / bibtex
	Trajectory Attention For Fine-grained Video Motion Control Zeqi Xiao, Wenqi Ouyang, Yifan Zhou, Shuai Yang, Lei Yang, Jianlou Si, Xingang Pan ICLR, 2025 project page / arXiv / code / bibtex
	Video Diffusion Models are Training-free Motion Interpreter and Controller Zeqi Xiao, Yifan Zhou, Shuai Yang, Xingang Pan NeurIPS, 2024 project page / arXiv / code / bibtex
	CooHOI: Learning Cooperative Human-Object Interaction with Manipulated Object Dynamics Jiawei Gao, Ziqin Wang, Zeqi Xiao, Jingbo Wang, Tai Wang, Jinkun Cao, Xiaolin Hu, Si Liu, Jifeng Dai, Jiangmiao Pang NeurIPS, 2024 (Spotlight) project page / arXiv / code / bibtex
	UniHSI: Unified Human-Scene Interaction via Prompted Chain-of-Contacts Zeqi Xiao, Tai Wang, Jingbo Wang, Jinkun Cao, Wenwei Zhang, Bo Dai, Dahua Lin, Jiangmiao Pang ICLR, 2024 (Spotlight) project page / arXiv / code / bibtex
	Position-Guided Point Cloud Panoptic Segmentation Transformer Zeqi Xiao, Wenwei Zhang, Tai Wang, Chen Change Loy, Dahua Lin, Jiangmiao Pang IJCV*, 2024 arXiv / code / bibtex

Design and source code from Jon Barron's website