Mike Zheng Shou

VideoMind: A Chain-of-LoRA Agent for Temporal-Grounded Video Reasoning

Draw-In-Mind: Rebalancing Designer-Painter Roles in Unified Multimodal Models Benefits Image Editing

TPDiff: Temporal Pyramid Video Diffusion Model

Paper2Video: Automatic Video Generation from Scientific Papers

D-AR: Diffusion via Autoregressive Models

DD-Ranking: Rethinking the Evaluation of Dataset Distillation

Personalized Vision via Visual In-Context Learning

UniMoD: Efficient Unified Multimodal Transformers with Mixture-of-Depths

A Gain for Reconstruction, A Pain for Generation: Exploiting Representation in Visual Tokenization

Ego-centric Predictive Model Conditioned on Hand Trajectories

UniRL: Self-Improving Unified Multimodal Models via Supervised and Reinforcement Learning

Automated Movie Generation via Multi-Agent CoT Planning

Rethinking Defense for Computer-Use Agents: Context Deception Attacks are Simple to Defend

MakeAnything: Harnessing Diffusion Transformers for Multi-Domain Procedural Sequence Generation

Code2Video: A Code-centric Paradigm for Educational Video Generation

Mitty: Diffusion-based Human-To-Robot Video Generation

DiffSeg30k: A Multi-Turn Diffusion Editing Benchmark for Localized AIGC Detection

WorldGUI: An Interactive Benchmark for Desktop GUI Automation from Any Starting Point

Computer-Use Agents as Judges for Automatic GUI Design

Anti-Reference: Universal and Immediate Defense Against Reference-Based Generation

Multi-Human Interactive Talking Dataset

macOSWorld: A Multilingual Interactive Benchmark for GUI Agents

Show-o2: Improved Native Unified Multimodal Models

Impossible Videos

Show-o: One Single Transformer to Unify Multimodal Understanding and Generation

PANDA: Towards Generalist Video Anomaly Detection via Agentic AI Engineer

Sparse Image Synthesis via Joint Latent and RoI Flow

OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data

Image Watermarks are Removable using Controllable Regeneration from Clean Noise

DOTA: Distributional Test-time Adaptation of Vision-Language Models

MP-Mat: A 3D-and-Instance-Aware Human Matting and Editing Framework with Multiplane Representation

CoFFT: Chain of Foresight-Focus Thought for Visual Language Models

Bridging Information Asymmetry in Text-video Retrieval: A Data-centric Approach

WMAdapter: Adding WaterMark Control to Latent Diffusion Models

Think or Not? Selective Reasoning via Reinforcement Learning for Vision-Language Models

Grounding Multimodal Large Language Model in GUI World

DOTA: Distributional Test-Time Adaptation of Vision-Language Models

Personalized Vision via Visual In-Context Learning

OmniContrast: Vision-Language-Interleaved Contrast from Pixels All at once

WMAdapter: Adding WaterMark Control to Latent Diffusion Models

VEditBench: Holistic Benchmark for Text-Guided Video Editing

Improving Autoregressive Image Generation by Mitigating Gradient Bias in Softmax

Unsupervised Prior Learning: Discovering Categorical Pose Priors from Videos

Anti-Reference: Universal and Immediate Defense Against Reference-Based Generation

X-PlugVid: Versatile Adaptation of Image Plugins for Controllable Video Generation