Wei Wu

18

Papers

142

Total Citations

Papers (18)

Language-Image Pre-training with Long Captions

Theoretical Benefit and Limitation of Diffusion Language Model

NeurIPS 2025arXiv

Benchmarking Large Vision-Language Models via Directed Scene Graph for Comprehensive Image Captioning

FreeDiff: Progressive Frequency Truncation for Image Editing with Diffusion Models

OmniCache: A Trajectory-Oriented Global Perspective on Training-Free Cache Reuse for Diffusion Transformer Models

RoboSense: Large-scale Dataset and Benchmark for Egocentric Robot Perception and Navigation in Crowded and Unstructured Environments

ProtCLIP: Function-Informed Protein Multi-Modal Learning

Learning Visual Generative Priors without Text

DriveScape: High-Resolution Driving Video Generation by Multi-View Feature Fusion

UniMamba: Unified Spatial-Channel Representation Learning with Group-Efficient Mamba for LiDAR-based 3D Object Detection

DynaAct: Large Language Model Reasoning with Dynamic Action Spaces

NeurIPS 2025arXiv

Causal Inference over Visual-Semantic-Aligned Graph for Image Classification

Synergy of GFlowNet and Protein Language Model Makes a Diverse Antibody Designer

HoloVIC: Large-scale Dataset and Benchmark for Multi-Sensor Holographic Intersection and Vehicle-Infrastructure Cooperative

SwiftPillars: High-Efficiency Pillar Encoder for Lidar-Based 3D Detection

GeoFormer: Geometry Point Encoder for 3D Object Detection with Graph-based Transformer

InstaDrive: Instance-Aware Driving World Models for Realistic and Consistent Video Generation

FreqPDE: Rethinking Positional Depth Embedding for Multi-View 3D Object Detection Transformers