CV
Junyan Lin
Ph.D. Student in Computer Science
linjyan00@163.com
+86-17367107694
Hong Kong, China,
Summary
Ph.D. student in Computer Science at The Hong Kong Polytechnic University. Research interests include Multimodal Large Language Models, Vision-Language-Action Models and World Models, and multisource remote sensing image classification.
Education
- Ph.D. Student in Computer Science2029The Hong Kong Polytechnic University, Department of ComputingCourses: Supervisors: Changwen Chen, Xin Jin
- Visiting Student2025Eastern Institute for Advanced Study, College of Information Science and TechnologyCourses: Supervisor: Xiaoyu Shen
- M.Phil. of Computer Science and Technology2025Ocean University of China, School of Computer Science and TechnologyGPA: 3.82/4.0Courses: Supervisor: Feng Gao
- B.S. of Computer Science and Technology2022Zhejiang Gongshang University, School of Computer Science and TechnologyGPA: 88/100 (Rank 1/68)
Skills
Programming Languages
- Python
- MATLAB
- C++
Publications
- Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language Models2026CVPR 2026 - IEEE/CVF Conference on Computer Vision and Pattern RecognitionJ. Zhang*, J. Tong*, Junyan Lin*, et al.
- Efficient Token Compression for the Understanding and Generation Unified MLLMs2026ISCAS 2026 - IEEE International Symposium on Circuits and SystemsJunyan Lin, J. Liu, S. Zhao, et al.
- Speak While Watching: Unleashing TRUE Real-Time Video Understanding Capability of Multimodal Large Language Models2026
- Multimodal Language Models See Better When They Look Shallower2025EMNLP 2025 - Conference on Empirical Methods in Natural Language ProcessingH. Chen*, Junyan Lin*, X. Chen, et al. Oral.
- Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices2025CVPR 2025 - IEEE/CVF Conference on Computer Vision and Pattern RecognitionJunyan Lin, H. Chen, Y. Fan, et al.
- Dynamic Cross-Modal Feature Interaction Network for Hyperspectral and LiDAR Data Classification2025IEEE Transactions on Geoscience and Remote SensingJunyan Lin, F. Gao, L. Qi, et al.
- Tell Codec What Worth Compressing: Semantically Disentangled Image Coding for Machine with LMMs2024VCIP 2024 - IEEE International Conference on Visual Communications and Image ProcessingJ. Liu, Y. Wei, Junyan Lin, et al.
- To Preserve or Compress: A Study of Connectors Through Perceptual Efficacy in Multimodal Large Language Models2024EMNLP 2024 - Conference on Empirical Methods in Natural Language ProcessingJunyan Lin, H. Chen, D. Zhu, et al.
- Hyperspectral Image Change Detection via Cross-Sample Slot Attention and Dual Gated Feed-Forward Network2024PRCV 2024 - Chinese Conference on Pattern Recognition and Computer VisionL. Cheng, Junyan Lin, F. Gao, et al.
- Boosting Spatial-Spectral Masked Auto-Encoder Through Mining Redundant Spectra for HSI-SAR/LiDAR Classification2024IGARSS 2024 - IEEE International Geoscience and Remote Sensing SymposiumJunyan Lin, X. Jin, F. Gao, et al.
- Sparse Focus Network for Multi-Source Remote Sensing Data Classification2024IGARSS 2024 - IEEE International Geoscience and Remote Sensing SymposiumX. Jin, Junyan Lin, F. Gao, et al.
- SS-MAE: Spatial-spectral masked autoencoder for multisource remote sensing image classification2023IEEE Transactions on Geoscience and Remote SensingJunyan Lin, F. Gao, X. Shi, et al.
- Gated-Cross Aggregation Network for Hyperspectral and LiDAR Data Classification2023IGARSS 2023 - IEEE International Geoscience and Remote Sensing SymposiumX. Shi, Junyan Lin, Y. Rao, et al.
- Hyperspectral and SAR Image Classification via Recursive Feature Interactive Fusion Network2023IGARSS 2023 - IEEE International Geoscience and Remote Sensing SymposiumL. Lv, Junyan Lin, F. Gao, et al.
- Multi-Scale Transformer Network for Hyperspectral Image Denoising2023IGARSS 2023 - IEEE International Geoscience and Remote Sensing SymposiumS. Hu, Y. Hu, Junyan Lin, et al.
Interests
- Interests and HobbiesMotorcycling, Fitness, Film Appreciation