Xiangyu YueEmail: xyyue [at] ie.cuhk.edu.hk |
|
I am currently an Assistant Professor in the Department of Information Engineering at Chinese University of Hong Kong, with the Multimedia Lab (MMLab).
I received my Ph.D. from Electrical Engineering and Computer Science at University of California, Berkeley, working with Prof. Alberto Sangiovanni Vincentelli and Prof. Kurt Keutzer at Berkeley AI Research. I am broadly interested in various areas including but not limited to: computer vision, multi-modal learning, generative models, foundation model, transfer learning, domain adaptation, interpretable systems, etc.
Prior to Berkeley, I received my M.S. degree from Stanford University and B.S. degree from Nanjing University. I have spent time at Google Research, Google [x] Robotics, Baidu AI Research, and Tencent AI Lab. I received the prestigious Lotfi A. Zadeh Award for my research work. *NEW* I have multiple fully-funded PhD (2025), MPhil, postdoc, RA, and intern positions available. Feel free to Email me if you are interested. Google Scholar | LinkedIn | Twitter | DBLP |
|
OneLLM: One Framework to Align All Modalities with Language
|
|
Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities
|
|
UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition
|
|
Meta-Transformer: A Unified Framework for Multimodal Learning
|
|
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model
|
|
Space Engage: Collaborative Space Supervision for Contrastive-based Semi-Supervised Semantic Segmentation
|
|
Beating Backdoor Attack at Its Own Game
|
|
Preventing Zero-Shot Transfer Degradation in Continual Learning of Vision-Language Models
|
|
Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained Models
|
|
MLSeg: Image and Video Segmentation as Multi-Label Classification and Selected-Label Pixel Classification
|
|
Conditional Synthetic Data Generation for Robust Machine Learning Applications with Limited Pandemic Data
|
|
Self-Supervised Pretraining Improves Self-Supervised Pretraining
|
|
Multi-source Few-shot Domain Adaptation
|
|
Unsupervised Point Cloud Pre-Training via View-Point Occlusion, Completion
|
|
On Ensemble Methods for Long-Tailed Recognition
|
|
AugPrune: Robust Network Pruning via Augmented Data
|
|
Scene-aware Learning Network for Radar Object Detection
|
|
Prototypical Cross-domain Self-supervised Learning for Few-shot Unsupervised Domain Adaptation
|
|
Curriculum Cyclegan for Textual Sentiment Domain Adaptation with Multiple Sources
|
|
Emotional Semantics-preserved and Feature-aligned CycleGAN for Visual Emotion Adaptation
|
|
A Review of Single-Source Deep Unsupervised Visual Domain Adaptation
|
|
PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation
|
|
Scenic: a Language for Scenario Specification and Scene Generation
|
|
Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization without Accessing Target Domain Data
|
|
Multi-source Domain Adaptation for Semantic Segmentation
|
|
Squeezesegv2: Improved model structure and unsupervised domain adaptation for road-object segmentation from a lidar point cloud
|
|
Counterexample-guided Data Augmentation
|
|
Shift: A Zero-flop, Zero-parameter Alternative to Apatial Convolutions
|
|
A LiDAR Point Cloud Generator: from a Virtual World to Autonomous Driving
|
|
Formal Specification for Deep Neural Networks
|
|
SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-time Road-object Segmentation from 3d LiDAR Point Cloud
|