Wei Hu

11

Papers

1

Total Citations

Papers (11)

What Happens During the Loss Plateau? Understanding Abrupt Learning in Transformers

NeurIPS 2025arXiv

HyperGCT: A Dynamic Hyper-GNN-Learned Geometric Constraint for 3D Registration

Large Scene Generation with Cube-Absorb Discrete Diffusion

Controllable Protein Sequence Generation with LLM Preference Optimization

Explicitly Perceiving and Preserving the Local Geometric Structures for 3D Point Cloud Attack

Knowledge Graph Error Detection with Contrastive Confidence Adaption

DHGCN: Dynamic Hop Graph Convolution Network for Self-Supervised Point Cloud Learning

LTGC: Long-tail Recognition via Leveraging LLMs-driven Generated Content

Generative 3D Part Assembly via Part-Whole-Hierarchy Message Passing

Seeing is Not Believing: Adversarial Natural Object Optimization for Hard-Label 3D Scene Attacks

DFlow: A Generative Model Combining Denoising AutoEncoder and Normalizing Flow for High Fidelity Waveform Generation