jiaming Zhou

Jiaming Zhou

I am a final-year PhD student at Hong Kong University of Science and Technology, Guangzhou, under the supervision of Prof. Junwei Liang.

Prior to this, I obtained my Bachelor's degree in Computer Science and Engineering, from Sichuan University in 2020. Then I received my Master degree in Computer Science and Engineering, at Sun Yat-Sen University in 2023, where I was advised by Prof. Wei-Shi Zheng.

My research focuses on generalizable robotic manipulation, particularly on investigating pre-training data and scalable model architectures for robot foundation models.

I am currently seeking a full-time position in industry. Please feel free to reach out if you know of a suitable opportunity.

Email / Scholar / Github

publications

Representative papers are highlighted.

	Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization Jiaming Zhou, Qihang Zhang, Gangwei Xu, Cunxin Fan, Yujie Zhao, Ruilin Wang, Yiming Luo, Shuai Yang, Xing Zhu, Yujun Shen, Junwei Liang, Yinghao Xu arXiv, 2026 [paper] [project page] [code] [post on X] An in-context world-action pre-training framework that uses human video prompts for zero-shot robotic task generalization.
	LingBot-VA 2.0: Native Video-Action Pretraining for Generalizable Robot Control Qihang Zhang, ..., Jiaming Zhou, ..., Yinghao Xu Technical Report, 2026 Core contributor on pre-training and pre-training data. [paper] [project page] A native video-action foundation model for generalizable robot control, featuring semantic visual-action tokenization, causal pre-training, a sparse MoE backbone, and asynchronous closed-loop inference.
	Exploring the Limits of Vision-Language-Action Manipulations in Cross-task Generalization Jiaming Zhou, Ke Ye, Jiayi Liu, Teli Ma, Zifan Wang, Ronghe Qiu, Kun-Yu Lin, Zhilin Zhao, Junwei Liang NeurIPS, 2025 [paper] [bibtex] [project page] A cross-task manipulation generalization benchmark to evaluate existing Vision-Language-Action (VLA) models and a novel generalizable VLA method.
	Mitigating the Human-Robot Domain Discrepancy in Visual Pre-training for Robotic Manipulation Jiaming Zhou, Teli Ma, Kun-Yu Lin, Zifan Wang, Ronghe Qiu, Junwei Liang CVPR, 2025 [paper] [bibtex] [project page] [量子位] A new paradigm utilizing paired human-robot videos to adapt human-data pretrained foundation models to robotic manipulation domain.
	Contrastive Imitation Learning for Language-guided Multi-Task Robotic Manipulation Teli Ma, Jiaming Zhou, Zifan Wang, Ronghe Qiu, Junwei Liang CoRL, 2024 [paper] [project page] [bibtex]
	GLOVER: Generalizable Open-Vocabulary Affordance Reasoning for Task-Oriented Grasping Teli Ma, Zifan Wang, Jiaming Zhou, Mengmeng Wang, Junwei Liang arxiv, 2025 [paper] [project page] [bibtex]
	EgoTraj-Bench: Towards Robust Trajectory Prediction Under Ego-view Noisy Observations Jiayi Liu, Jiaming Zhou, Ke Ye, Kun-Yu Lin, Allan Wang, Junwei Liang ICRA, 2026 [paper] [bibtex]
	From Watch to Imagine Steering Long-horizon Manipulation via Human Demonstration and Future Envisionment Ke Ye, Jiaming Zhou (co-first), Yuanfeng Qiu, Jiayi Liu, Shihui Zhou, Kun-Yu Lin, Junwei Liang arxiv, 2025 [paper] [bibtex] [project page] [深蓝具身智能] A zero-shot long-horizon manipulation framework that mimics human long-range activities via demonstrations and achieves robust execution via generating visual future.
	Omni-Perception: Omnidirectional Collision Avoidance for Legged Locomotion in Dynamic Environments Zifan Wang, Teli Ma, Yufei Jia, Xun Yang, Jiaming Zhou, Wenlong Ouyang, Qiang Zhang, Junwei Liang CoRL, 2025 [paper] [project page] [bibtex]
	GLOVER++: Enhanced Generalizable Open-Vocabulary Affordance Reasoning for Task-Oriented Grasping Teli Ma, Jia Zheng, Zifan Wang, Ziyao Gao, Jiaming Zhou, Junwei Liang CoRL, 2025 [paper] [project page] [bibtex]
	AdaFocus: Towards End-to-end Weakly Supervised Learning for Long-Video Action Understanding Jiaming Zhou, Hanjun Li, Kun-Yu Lin, Junwei Liang arxiv, 2024 [paper] [bibtex] [project page] The first weakly supervised framework for developing efficient end-to-end action recognition models on long videos, which gives birth to a new weakly supervised pipeline for downstream long-video tasks.
	ActionHub: A Large-scale Action Video Description Dataset for Zero-shot Action Recognition Jiaming Zhou, Junwei Liang, Kun-Yu Lin, Jinrui Yang, Wei-Shi Zheng arxiv, 2024 [paper] [bibtex] A large-scale action video description dataset named ActionHub is proposed, which is the first, and the largest dataset that provides millions of video descriptions to describe thousands of human actions.
	TwinFormer: Fine-to-Coarse Temporal Modeling for Long-term Action Recognition Jiaming Zhou, Kun-Yu Lin, Yu-Kun Qiu, Wei-Shi Zheng Transactions on Multimedia (TMM), 2023 [paper] [bibtex]
	Graph-based High-order Relation Modeling for Long-term Action Recognition Jiaming Zhou, Kun-Yu Lin, Haoxin Li, Wei-Shi Zheng CVPR, 2021 [paper] [supp] [bibtex]
	Adversarial Partial Domain Adaptation by Cycle Inconsistency Kun-Yu Lin, Jiaming Zhou (co-first), Yu-Kun Qiu, Wei-Shi Zheng ECCV, 2022 [paper] [supp] [bibtex]
	Human-Centric Transformer for Domain Adaptive Action Recognition Kun-Yu Lin, Jiaming Zhou, Wei-Shi Zheng TPAMI, 2024 [paper]
	XOV-Action: Towards Generalizable Open-Vocabulary Action Recognition Kun-Yu Lin, Henghui Ding, Jia-Run Du, Jiaming Zhou, Yi-Xing Peng, Zhilin Zhao, Chen Change Loy, Wei-Shi Zheng TPAMI, 2026 [paper] [bibtex] [github] The first benchmark, named XOV-Action, for the cross-domain open-vocabulary action recognition task, and a simple yet effective method to address the scene bias for the task.
	GeoDeformer: Geometric Deformable Transformer for Action Recognition Jinhui Ye, Jiaming Zhou, Hui Xiong, Junwei Liang arxiv, 2023 [paper] [bibtex]
	Diversifying Spatial-Temporal Perception for Video Domain Generalization Kun-Yu Lin, Jia-Run Du, Yipeng Gao, Jiaming Zhou, Wei-Shi Zheng NeurIPS, 2023 [paper] [supp] [bibtex] [github]