Chenfeng Xu

QuantV2X: A Fully Quantized Multi-Agent System for Cooperative Perception

REAR: Scalable Test-time Preference Realignment through Reward Decomposition

Angles Don’t Lie: Unlocking Training‑Efficient RL Through the Model’s Own Signals

Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation

WOMD-Reasoning: A Large-Scale Dataset for Interaction Reasoning in Driving

Looking Backward: Streaming Video-to-Video Translation with Feature Banks

Law of Vision Representation in MLLMs

Dobi-SVD: Differentiable SVD for LLM Compression and Some New Perspectives

Sparse Video-Gen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity

X-Drive: Cross-modality Consistent Multi-Sensor Data Synthesis for Driving Scenarios

Law of Vision Representation in MLLMs

CompGS: Unleashing 2D Compositionality for Compositional Text-to-3D via Dynamically Optimizing 3D Gaussians