Xiaodan Liang

SemHiTok: A Unified Image Tokenizer via Semantic-Guided Hierarchical Codebook for Multimodal Understanding and Generation

iTryOn: Mastering Interactive Video Virtual Try-On with Spatial-Semantic Guidance

FastFit: Accelerating Multi-Reference Virtual Try-On via Cacheable Diffusion Models

Does Your 3D Encoder Really Work? A simple yet effective pathway to real 3D scene understanding

C2-Evo: Co-Evolving Multimodal Data and Model for Self-Improving Reasoning

VistaGUI: Towards More Robust and Intelligent GUI Automation

SimuPhy: Towards Physical Understanding, Reasoning, and Evaluation via Code Generation

Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration

Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers

MakeupAnyone: Self-Supervised Identity-Preserving MakeUp Transfer with Region-Aware Multi-Scale Alignment

TreeRPO: Tree Relative Policy Optimization

CombiBench: Benchmarking LLM Capability for Combinatorial Mathematics

SPC: Evolving Self-Play Critic via Adversarial Games for LLM Reasoning

MMTryon: Multi-Modal Multi-Reference Control for High-Quality Fashion Generation

WISA: World simulator assistant for physics-aware text-to-video generation

OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling

GDrag:Towards General-Purpose Interactive Editing with Anti-ambiguity Point Diffusion

PT-T2I/V: An Efficient Proxy-Tokenized Diffusion Transformer for Text-to-Image/Video-Task

CatVTON: Concatenation Is All You Need for Virtual Try-On with Diffusion Models

UniGS: Unified Language-Image-3D Pretraining with Gaussian Splatting

Sitcom-Crafter: A Plot-Driven Human Motion Generation System in 3D Scenes

S2-Track: A Simple yet Strong Approach for End-to-End 3D Multi-Object Tracking

EMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions

UncertaintyRAG: Span Uncertainty Enhanced Long-Context Modeling for Retrieval-Augmented Generation

Continual LLaVA: Continual Instruction Tuning in Large Vision-Language Models

HiRes-LLaVA: Restoring Fragmentation Input in High-Resolution Large Vision-Language Models

StoryAgent: Customized Storytelling Video Generation via Multi-Agent Collaboration

Memory-Driven Multimodal Chain of Thought for Embodied Long-Horizon Task Planning

ActionFiller: Fill-In-The-Blank Prompting for OS Agent