Siyuan Li (李思远) |
Selected Preprints:
Unveiling the Backbone-Optimizer Coupling Bias in Visual Representation Learning |
Switch EMA: A Free Lunch for Better Flatness and Sharpness |
Conferences (As First Author):
VQDNA: Unleashing the Power of Vector Quantization for Multi-Species Genomic Sequence Modeling |
SemiReward: A General Reward Model for Semi-supervised Learning |
MogaNet: Efficient Multi-order Gated Aggregation Network |
OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning |
Harnessing Hard Mixed Samples with Decoupled Regularizer |
Architecture-Agnostic Masked Image Modeling -- From ViT back to CNN |
Exploring Localization for Self-supervised Fine-grained Contrastive Learning |
AutoMix: Unveiling the Power of Mixup for Stronger Classifiers |
DLME: Deep Local-flatness Manifold Embedding |
Invertible Manifold Learning for Dimension Reduction |
TLPG-Tracker: Joint Learning of Target Localization and Proposal Generation for Visual Tracking |
Conferences (As Co-author):
Boosting the Power of Small Multimodal Reasoning Models to Match Larger Models with Self-Consistency Training |
Short-Long Convolutions Help Hardware-Efficient Linear Attention to Focus on Long Sequences |
Re-Dock: Towards Flexible and Realistic Molecular Docking with Diffusion Bridge |
Learning to Predict Mutation Effects of Protein-Protein Interactions by Microenvironment-aware Hierarchical Prompt Learning |
LongVQ: Long Sequence Modeling with Vector Quantization on Structured Memory |
MAPE-PPI: Towards Effective and Efficient Protein-Protein Interaction Prediction via Microenvironment-Aware Protein Embedding |
Hierarchical Data-efficient Representation Learning for Tertiary Structure-based RNA Design |
KW-Design: Pushing the Limit of Protein Deign via Knowledge Refinement |
Protein 3D Graph Structure Learning for Robust Structure-based Protein Property Prediction |
Wavelet-Driven Spatiotemporal Predictive Learning: Bridging Frequency and Time Variations |
Functional-Group-Based Diffusion for Pocket-Specific Molecule Generation and Elaboration |
Explaining Graph Neural Networks via Non-parametric Subgraph Matching |
CVT-SLR: Contrastive Visual-Textual Transformation for Sign Language Recognition with Variational Alignment |
Temporal attention unit: Towards efficient spatiotemporal predictive learning |
Mole-BERT: Rethinking Pre-training Graph Neural Networks for Molecules |
Divide and Contrast: Source-free Domain Adaptation via Adaptive Contrastive Learning |
Are Gradients on Graph Structure Reliable in Gray-box Attacks? |
Hyperspherical Consistency Regularization |
Generalized Clustering and Multi-Manifold Learning with Geometric Structure Preservation |
Journals:
GenURL: A General Framework for Unsupervised Representation Learning |
Discovering the Representation Bottleneck of Graph Neural Networks |
Neuro-BERT: Rethinking Masked Autoencoding for Self-Supervised Neurological Pretraining |
GNN Cleaner: Label Cleaner for Graph-structured Data |
Deep Manifold Attributed Graph Embedding |
Membership:
Invited Talk:
Program Committee Member | Reviewer:
Teaching Assistant:
Online Project:
Social Awards: