Chao Ma

PointLAM: Local Attentive Mamba for Efficient Point-based 3D Object Detection

P-LoRA: Posterior Knowledge Enables Training-Free Fusion of Subject and Style LoRAs

Tokens that Know Where: Self-improving 2D Spatial Vocabulary for Multi-modal Understanding

UNIFIED MULTI-TEACHER DISTILLATION ACROSS HYBRID NEURAL ARCHITECTURES

PhyCAGE: Physically Constrained Compositional 3D Asset Generation from a Single Image

SimTrack3D: A Simple Sequential Motion Modeling for Efficient 3D Single Object Tracking

GaussUnveil: Unified Occlusion-Aware Gaussian Refinement for Sparse-View Scene Reconstruction

BachVid: Training-Free Video Generation with Consistent Background and Character

ImageDriver: Let Vision-Language-Action Models Drive on 2D Images

IPG-Rec: Instance-aware Progressive Geometry Rectification for High-fidelity Sparse Novel View Synthesis

Co-Reinforcement Learning for Unified Multimodal Understanding and Generation

A Simple Approach to Unifying Diffusion-based Conditional Generation

Correlated Low-Rank Adaptation for ConvNets

Space-Correlated Transformer: Jointly Explore the Matching and Motion Clues in 3D Single Object Tracking

VRM: Knowledge Distillation via Virtual Relation Matching

PredFormer: Transformers Are Effective Spatial-Temporal Predictive Learners

Invisibility Stickers Against LiDAR: Adversarial Attacks on Point Cloud Intensity for LiDAR-based Detection