Muzhi Zhu

NoTVLA: Narrowing of Dense Action Trajectories for Generalizable Robot Manipulation

ACTIVE-o3 : Empowering MLLMs with Active Perception via Pure Reinforcement Learning

Beyond Identity: High-Fidelity Face Swapping by Preserving Source Video Attributes

HieraTok: Multi-Scale Visual Tokenizer Improves Image Reconstruction and Generation

Bridge Thinking and Acting: Unleashing Physical Potential of VLM with Generalizable Action Expert

GUI-Shepherd: Reliable Process Reward and Verification for Long-Sequence GUI Tasks

DICEPTION: A Generalist Diffusion Model for Visual Perceptual Tasks

Omni-R1: Reinforcement Learning for Omnimodal Reasoning via Two-System Collaboration

Bridging the Gap between Semantic Correspondence and Robust Visual Representation