Cheng Shi (石骋) PhD Student Department of Computer Science The University of Hong Kong Email: shicheng2025 [at] connect [dot] hku [dot] hk GitHub | Google Scholar

Biography

I am currently a first-year Ph.D. student at Department of Computer Science, The University of Hong Kong, where I have the privilege of being supervised by Prof. Yizhou Yu (ACM/IEEE Fellow). Previously, I obtained my Master's degree from ShanghaiTech University in 2024, where I was advised by Prof. Sibei Yang, and my Bachelor's degree from ShanghaiTech University in 2022.

My research interests lie at the intersection of computer vision, natural language processing, and multimodal AI. My current research focuses on open-world visual perception and vision foundation models.

News

Selected Publications

* denotes equal contribution and † corresponding author

	Vision Transformer Needs More Than Register In submission, 2025 Cheng Shi and Sibei Yang† Paper
	Vision Function Layer in Multimodal LLMs NeurIPS 2025 Cheng Shi, Yizhou Yu and Sibei Yang Paper Code
	Eyes Wide Open: Ego Proactive Video-LLM for Streaming Video NeurIPS 2025 Yulin Zhang, Cheng Shi, Yang Wang and Sibei Yang† Paper
	Discovering Compositional Hallucination in LVLMs NeurIPS 2025 Ge Zheng, Jiajin Tang, Jiaye Qian, Hanzhuo Huang, Cheng Shi and Sibei Yang† Paper
	Part2Object: Hierarchical Unsupervised 3D Instance Segmentation ECCV 2024 Cheng Shi*, Yuling Zhang*, Bin Yang, Jiajin Tang, Yuexin Ma and Sibei Yang† Paper Code
	Plain-DNet: A Plain Multi-Dataset Object Detector ECCV 2024 Cheng Shi*, Yuchen Zhu* and Sibei Yang† Paper Code
	Zip-Your-CLIP: CLIP Itself is a Good Object-detector ICLR 2024 Cheng Shi and Sibei Yang† Paper Code
	Free-Bloom: Zero-Shot Text-to-Video Generator with LLM Director and LDM Animator NeurIPS 2023 Hanzhuo Huang*, Yufan Feng*, Cheng Shi, Lan Xu, Jingyi Yu, and Sibei Yang† Paper Code
	LoGoPrompt: Synthetic Text Images Can Be Good Visual Prompts for Vision-Language Models ICCV 2023 Cheng Shi, and Sibei Yang† Project Paper
	EdaDet: Open-Vocabulary Object Detection Using Early Dense Alignment ICCV 2023 Cheng Shi, and Sibei Yang† Project Paper
	Contrastive Grouping with Transformer for Referring Image Segmentation CVPR 2023 Jiajin Tang, Ge Zheng, Cheng Shi, and Sibei Yang† Paper Code
	DreamFace: Progressive Generation of Animatable 3D Faces under Text Guidance SIGGRAPH 2023 Longwen Zhang*, Qiwei Qiu*, Hongyang Lin*, Qixuan Zhang, Cheng Shi, Wei Yang, Ye Shi, Sibei Yang†, Lan Xu†, Jingyi Yu† Project Paper Video
	Spatial and Visual Perspective-Taking via View Rotation and Relation Reasoning for Embodied Reference Understanding ECCV 2022 Cheng Shi, and Sibei Yang† Paper Code