Yunchao Wei

Semantic Visual Anomaly Detection and Reasoning in AI-Generated Images

Matting Anything 2: Towards Video Matting for Anything

PanoWorld-X: Generating Explorable Panoramic Worlds via Sphere-Aware Video Diffusion

EvoProto: Evolving Prototypes with Class Similarity for Weakly Incremental Segmentation

CoCoDiff: Correspondence-Consistent Diffusion Model for Fine-grained Style Transfer

SpatialTree: Branching Out Spatial Intelligence in MLLMs via a Capability Tree

InsertAny3D: VLM-Assisted and Geometry-Grounded Framework for 3D Object Insertion in Complex 3D Scenes

Adaptive Sensor Selection for Power Efficient 3D Object Detection on Autonomous Driving

M4V: Multimodal Mamba for Efficient Text-to-Video Generation

Hybrid Neural-MPM for Interactive Fluid Simulations in Real-Time

PreFM: Online Audio-Visual Event Parsing via Predictive Future Modeling

Collapsed Language Models Promote Fairness

IPSeg: Image Posterior Mitigates Semantic Drift in Class-Incremental Segmentation

DCI: Dual-Conditional Inversion for Boosting Diffusion-Based Image Editing

A Unified Reasoning Framework for Holistic Zero-Shot Video Anomaly Analysis

ClassDiffusion: More Aligned Personalization Tuning with Explicit Class Guidance

Intent3D: 3D Object Detection in RGB-D Scans Based on Human Intention

FlexVAR: Flexible Visual Autoregressive Modeling without Residual Prediction