Jinlong Li

Jinlong Li

PhD Student, University of Trento (MHUG)
Visiting PhD Student, ETH Zurich (CVG)

Email: jinlong.szu@gmail.com; jinlong.li@unitn.it

Pushing a long-termist, strike the tough yet right things correctly.

Biography

I am a visiting PhD student at the Computer Vision and Geometry Group (CVG), ETH Zurich, Switzerland, under the supervision of Prof. Marc Pollefeys, and a PhD student at the Multimedia and Human Understanding Group (MHUG), University of Trento, Italy, under the supervision of Prof. Nicu Sebe. Previously, I was a research scientist in computer vision at Meituan, working closely with Dr. Lin Ma and Dr. Zequn Jie on label-efficient 2D/3D/4D detection and segmentation. Before that, I received a dual B.Sc. degree (major: Applied Physics) and an M.Sc. degree in Computer Science from Shenzhen University.

Research Interest

I work on computer vision and deep learning. My recent research aims to build machines that can see, reason about, and interact with the geometric and physical world through a unified end-to-end paradigm built on representation and generative learning. Topics I am exploring include:

Multi-Modal Perception, Reasoning and Planning
Spatial Foundation Models
Generative Models
Label-Efficient Learning in 2D, 3D and 4D

News

2026.09: Two papers, OneVL and LECDrive (VLA-based autonomous driving), got accepted by NeurIPS 2026! Congrats to all co-authors!
2026.07: One paper, Stable-ST, got accepted by IJCV 2026! Congrats to Dong!
2026.05: We released OneVL, the first latent CoT method to surpass explicit CoT in autonomous driving trajectory planning while matching answer-only inference latency, via a unified VLA and World Model framework.
2026.02: Our work AOT got accepted by CVPR 2026! Congrats to all co-authors!
2025.02: Our works CUA_O3D and FisherTune got accepted by CVPR 2025! Congrats to all co-authors!
2024.09: Our work LESS got accepted by NeurIPS 2024!
2024.07: Our work RKP got accepted by ACM MM 2024!
2024.07: Our work 3DSS-VLG got accepted by ECCV 2024!

Earlier news

2023.02: One paper got accepted by Neurocomputing 2023!
2022.09: Our work ESOL got accepted by NeurIPS 2022!
2022.02: Our work PPL got accepted by TMM 2022!

PrePrints

	A Survey on Post-training of Multimodal Large Language Models Haonan Zhang, Pengpeng Zeng, Libin Cao, Wenrui Lai, Jinlong Li, Duo Peng, Yi Bin, Xuanhan Wang, Ji Zhang, Jingkuan Song, Nicu Sebe, Yuchuan Wu, Yongbin Li, Heng Tao Shen, Jieping Ye. [Preprint] [BibTeX]
	3D Weakly Supervised Semantic Segmentation via Class-Aware and Geometry-Guided Pseudo-Label Refinement Xiaoxu Xu, Xuexun Liu, Jinlong Li, Yitian Yuan, Qiudan Zhang, Lin Ma, Nicu Sebe, Xu Wang. [Arxiv] [BibTeX]
	Orthogonal Projection Subspace to Aggregate Online Prior-knowledge for Continual Test-time Adaptation Jinlong Li, Dong Zhao, Qi Zang, Zequn Jie, Lin Ma, Nicu Sebe. [Arxiv] [BibTeX]
	Feature Compression for Vision-Language Models: Query-Driven Encoding in Bandwidth-Constrained Environments Pingping Zhang, Jinlong Li, Kecheng Chen, Meng Wang, Yuzhi Zhao, Haoliang Li, Sam Kwong, Shiqi Wang. [Arxiv]
	Enhancing Robustness of Vision-Language Models through Orthogonality Learning and self-Regularization Jinlong Li, Zequn Jie, Elisa Ricci, Lin Ma, Nicu Sebe. [Arxiv] [BibTeX]

Publications

(* equal contribution, † corresponding author)

	Think Densely, Act Sparsely: Latent Expert Cognitive Chains for Vision-Language-Action Autonomous Driving Jie Wang, Guang Li, Zhijian Huang, Jinlong Li, Chenxu Dang, Hangjun Ye, Yahong Han, Long Chen. Neural Information Processing Systems (NeurIPS), 2026. [Project] [Code] [Arxiv] [BibTeX]
	OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation Jinghui Lu, Jiayi Guan, Zhijian Huang, Jinlong Li, Guang Li, Lingdong Kong, et al. Neural Information Processing Systems (NeurIPS), 2026. [Project] [Technical Report] [BibTeX] [Core Contributor (PhysicalAI-Autonomous-Vehicles Develop)]
	Towards Stable Source-Free Domain Adaptive Semantic Segmentation Dong Zhao, Qi Zang, Nan Pu, Jinlong Li, Shuang Wang, Nicu Sebe, Zhun Zhong. International Journal of Computer Vision (IJCV), 2026. [Paper Link] [Code] [BibTeX]
	Token Reduction via Local and Global Contexts Optimization for Efficient Video Large Language Models Jinlong Li, Liyuan Jiang, Haonan Zhang, Nicu Sebe. Conference on Computer Vision and Pattern Recognition (CVPR), 2026. [Project] [ArXiv Link] [Code] [BibTeX]
	When Video Coding Meets Multimodal Large Language Models: A Unified Paradigm for Video Coding Pingping Zhang, Jinlong Li, Meng Wang, Nicu Sebe, Sam Kwong, Shiqi Wang. IEEE Signal Processing Letters (SPL), 2026. [ArXiv Link] [BibTeX]
	Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understanding Jinlong Li, Cristiano Saltori, Fabio Poiesi, Nicu Sebe. Conference on Computer Vision and Pattern Recognition (CVPR), 2025. [Project] [ArXiv Link] [Code] [BibTeX]
	FisherTune: Fisher-Guided Robust Tuning of Vision Foundation Models for Domain Generalized Segmentation Dong Zhao, Jinlong Li, Shuang Wang, Mengyao Wu, Qi Zang, Nicu Sebe, Zhun Zhong. Conference on Computer Vision and Pattern Recognition (CVPR), 2025. [ArXiv Link] [Code] [BibTeX]
	LESS: Label-Efficient and Single-Stage Referring 3D Instance Segmentation Xuexun Liu, Xu Xiaoxu, Jinlong Li, Qiudan Zhang, Xu Wang, Lin Ma, Nicu Sebe. Neural Information Processing Systems (NeurIPS*), 2024. [ArXiv Link] [Code] [BibTeX]
	Generalized Source-free Domain-adaptive Segmentation via Reliable Knowledge Propagation Qi Zang, Shuang Wang, Dong Zhao, Yang HU, Dou Quan, Jinlong Li, Nicu Sebe, Zhun Zhong. ACM Multimedia (ACM MM), 2024. [Paper Link] [BibTeX]
	3D Weakly Supervised Semantic Segmentation with 2D Vision-Language Guidance Xiaoxu Xu, Yitian Yuan, Jinlong Li, Qiudan Zhang, Zequn Jie, Lin Ma, Hao Tang, Nicu Sebe, Xu Wang. European Conference on Computer Vision (ECCV), 2024. [ArXiv Link] [Code] [BibTeX]
	Weakly Supervised Semantic Segmentation via Self-Supervised Destruction Learning Jinlong Li, Zequn Jie, Xu Wang, Yu Zhou, Lin Ma, Jianming Jiang. Neurocomputing (NEUCOM), 2023. [BibTeX]
	Expansion and Shrinkage of Localization for Weakly-Supervised Semantic Segmentation Jinlong Li, Zequn Jie, Xu Wang, Xiaolin Wei, Lin Ma. Neural Information Processing Systems (NeurIPS), Spotlight (1.7%) 2022. [ArXiv Link] [Code] [BibTeX]
	Weakly Supervised Semantic Segmentation via Progressive Patch Learning Jinlong Li, Zequn Jie, Xu Wang, Yu Zhou, Xiaolin Wei, Lin Ma. IEEE Transactions on Multimedia (TMM), 2022. [ArXiv Link] [IEEE Trans Link] [Code] [BibTeX]