Shaofeng Zhang

Dual-Branch Center-Surrounding Contrast: Rethinking Contrastive Learning for 3D Point Clouds

Many Can Beat One: MoE-Linear Attention for Full MLP Image Generation

PointARSim: Point-cloud–enhanced Generative Auto-Regressive Simulation for Closed-loop End-to-end Autonomous Driving Evaluation

Understanding Data Influence in Reinforcement Finetuning

VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models

DLoFT: Gradient-Decoupled Fine-Tuning for Generalizable Long Chain-of-Thought Reasoning

Position-Query-Based Autoencoders for View Decoupled Cross Point Cloud Reconstruction and a Self-Supervised Learning Framework

CR2PQ: Continuous Relative Rotary Positional Query for Dense Visual Representation Learning

Learn to Synthesize Compact Datasets by Matching Effects