Zedong Wang (王泽栋) |
Selected Preprints (*: Equal Contribution. †: Corresponding Author.):
Switch EMA: A Free Lunch for Better Flatness and Sharpness |
Masked Modeling for Self-supervised Representation Learning on Vision and Beyond |
OpenMixup: A Comprehensive Mixup Benchmark for Visual Classification |
Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup |
Conferences (As First Author):
VQDNA: Unleashing the Power of Vector Quantization for Multi-Species Genomic Sequence Modeling |
MogaNet: Multi-order Gated Aggregation Network |
Conferences (As Co-author):
Short-Long Convolutions Help Hardware-Efficient Linear Attention to Focus on Long Sequences |
LongVQ: Long Sequence Modeling with Vector Quantization on Structured Memory |
SemiReward: A General Reward Model for Semi-supervised Learning |
OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning |
Program committee member | Reviewer:
Membership: