Ziwei Liu

EgoTwin: Dreaming Body and View in First Person

IGGT: Instance-Grounded Geometry Transformer for Semantic 3D Reconstruction

JavisDiT: Joint Audio-Video Diffusion Transformer with Hierarchical Spatio-Temporal Prior Synchronization

From Spatial to Actions: Grounding Vision-Language-Action Model in Spatial Foundation Priors

Light-X: Generative 4D Video Rendering with Camera and Illumination Control

From Pixels to Words -- Towards Native Vision-Language Primitives at Scale

EchoBench: Benchmarking Sycophancy in Medical Large Vision-Language Models

VISA: Preserving Fine-Grained Perception in MLLMs via Visual Semantic Anchoring

JavisDiT++: Unified Modeling and Optimization for Joint Audio-Video Generation

The Quest for Generalizable Motion Generation: Data, Model, and Evaluation

Visual Jigsaw Post-Training Improves MLLMs

4DNeX: Feed-Forward 4D Generative Modeling Made Easy

Ego-R1: Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoning

ArtHOI: Articulated Human-Object Interaction Synthesis via Dynamics Distillation

RealDPO: Real or Not Real, that is the Preference

Demo-ICL: In-Context Learning for Procedural Video Knowledge Acquisition

RealUnify: Do Unified Models Truly Benefit from Unification? A Comprehensive Benchmark

HSImul3R: Reconstructing Simulation-Ready Human-Scene-Interaction from Sparse Views

SafeRBench: A Comprehensive Benchmark for Safety Assessment of Large Reasoning Models

DiverseAR: Boosting Diversity in Bitwise Autoregressive Image Generation

PhysX-3D: Physical-Grounded 3D Asset Generation

Video World Models with Long-term Spatial Memory

DynamicCity: Large-Scale 4D Occupancy Generation from Dynamic Scenes

ShotBench: Expert-Level Cinematic Understanding in Vision-Language Models

Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos

Imagine360: Immersive 360 Video Generation from Perspective Anchor

VideoLucy: Deep Memory Backtracking for Long Video Understanding

GOOD: Training-Free Guided Diffusion Sampling for Out-of-Distribution Detection

GUI-Reflection: Empowering Multimodal GUI Models with Self-Reflection Behavior

GeneMAN: Generalizable Single-Image 3D Human Reconstruction from Multi-Source Human Data

Phidias: A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion

Oryx MLLM: On-Demand Spatial-Temporal Understanding at Arbitrary Resolution

Long Context Transfer from Language to Vision

FreeMorph: Tuning-Free Generalized Image Morphing with Diffusion Model

Unsolvable Problem Detection: Evaluating Trustworthiness of Large Multimodal Models

InterDance: Reactive 3D Dance Generation with Realistic Duet Interactions

GS-VTON: Controllable 3D Virtual Try-on with Gaussian Splatting

AvatarGO: Zero-shot 4D Human-Object Interaction Generation and Animation

FasterCache: Training-Free Video Diffusion Model Acceleration with High Quality

FreeTraj: Tuning-Free Trajectory Control via Noise Guided Video Diffusion

EgoLM: Multi-Modal Language Model of Egocentric Motions

VEnhancer: Generative Space-Time Enhancement for Video Generation

3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion

Video Instruction Tuning with Synthetic Data

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

Streamline Without Sacrifice - Squeeze out Computation Redundancy in LMM

MVPaint: 3D Texture Generation with Multi-View Consistency