Liu He

I am an applied scientist at Amazon, working on generative video/image synthesis, and multimodal reasoning in-the-wild. I received my CS PhD at Purdue University, under Daniel Aliaga. My passion is to let AI know the physical world. My research focusing on synthetic data generation, advancing MLLM reasoning on visual tasks, and physcial-aware VideoGen as World Model. I am all-time open to collaborations. Contact me any time :-)

I am the Cover Man on 2025 Indiana State Fishing Guide, see my face in any outdoor stores in Indiana :-)

Email / CV / GitHub / Google Scholar / LinkedIn

Research

	PhysAlign: Physics-Coherent Video Generation through Feature and 3D Representation Alignment Zhexiao Xiong, Yizhi Song, Liu He, Wei Xiong, Yu Yuan, Feng Qiao, Nathan Jacobs Under Review Project / Paper
	Advancing MLLMs by Large-Scale 3D Visual Instruction Dataset Generation Liu He, Xiao Zeng, Yizhi Song, Albert Y. C. Chen, Lu Xia, Shashwat Verma, Sankalp Dayal, Min Sun, Daniel Aliaga WACV 2026 Oral \| Award Finalist Project / Paper
	DocAgent: An Agentic Framework for Multi-Modal Long-Context Document Understanding Li Sun, Liu He, Shuyue Jia, Yangfan He, Chenyu You EMNLP 2025 Main Paper
	LongPerceptualThoughts: Distilling System-2 Reasoning for System-1 Perception Yuan-Hong Liao, Sven Elflein, Liu He, Laura Leal-Taixé, Yejin Choi, Sanja Fidler, David Acuna COLM 2025 Project / Paper / Code / Dataset
	MMIG-Bench: Towards Comprehensive and Explainable Evaluation of Multi-Modal Image Generation Models Hang Hua, Ziyun Zeng, Yizhi Song, Yunlong Tang, Liu He, Daniel Aliaga, Wei Xiong, Jiebo Luo NeurIPS 2025 Project / Paper / Code / Data
	Kubrick: Multimodal Agent Collaborations for Synthetic Video Generation Liu He, Yizhi Song, Yunlong Tang, Pinxin Liu, Hejun Huang, Daniel Aliaga, Xin Zhou CVPR 2025 AI4CC Workshop Project / Paper
	Refine-by-Align: Reference-Guided Artifacts Refinement through Semantic Alignment Yizhi Song, Liu He, Zhifei Zhang, Soo Ye Kim, He Zhang, Wei Xiong, Zhe Lin, Brian Price, Scott Cohen, Jianming Zhang, Daniel Aliaga ICLR 2025 Project / Paper
	COHO: Context-Sensitive City-Scale Hierarchical Urban Layout Generation Liu He, Daniel Aliaga ECCV 2024 Oral Project / Paper / Code / Dataset
	GlobalMapper: Arbitrary-Shaped Urban Layout Generation Liu He, Daniel Aliaga ICCV 2023 Project / Paper / Supp / Code
	Diffusion-Based Document Layout Generation Liu He, Yijuan Lu, John Corring, Dinei Florencio, Cha Zhang ICDAR 2023 Oral Project / Paper / Code
	Generative Building Feature Estimation from Satellite Images Liu He, Jie Shan, Daniel Aliaga IEEE Transactions on Geoscience and Remote Sensing (2023) Project / Paper
	Deep Learning-Based Urban Morphology for City-Scale Environmental Modeling Pratiman Patel, Rajesh Kalyanam, Liu He, Daniel Aliaga, Dev Niyogi PNAS Nexus (2023) Project / Paper