Xihui Liu

HoloPart: Generative 3D Part Amodal Segmentation

Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-Language Navigation

AnimeShooter: A Multi-Shot Animation Dataset for Reference-Guided Video Generation

FLUX-Reason-6M & PRISM-Bench: A Million-Scale Text-to-Image Reasoning Dataset and Comprehensive Benchmark

Routing Matters in MoE: Scaling Diffusion Transformers with Explicit Routing Guidance

SRUM: Fine-Grained Self-Rewarding for Unified Multimodal Models

OmniX: From Unified Panoramic Generation and Perception to Graphics-Ready 3D Scenes

GoT-R1: Unleashing Reasoning Capability of Autoregressive Visual Generation with Reinforcement Learning

IPCP: Interpreter, Planner, Checker, and Painter Dialogue for Compositional Text-to-Image Generation

FilMaster: Bridging Cinematic Principles and Generative AI for Automated Film Generation

ChemBOMAS: Accelerated Bayesian Optimization for Scientific Discovery in Chemistry with LLM-Enhanced Multi-Agent System

T2I-ReasonBench: Benchmarking Reasoning-Informed Text-to-Image Generation

GRPO-CARE: Consistency-Aware Reinforcement Learning for Multimodal Reasoning

Learning Interpretable Representations Leads to Semantically Faithful EEG-to-Text Generation

CodePlot-CoT: Mathematical Visual Reasoning by Thinking with Code-Driven Images

Building Scalable Real-World Robot Data Generation via Compositional Simulation

Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance

Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generation

UniMC: Taming Diffusion Transformer for Unified Keypoint-Guided Multi-Class Image Generation

Understand Before You Generate: Self-Guided Training for Autoregressive Image Generation

TTS-VAR: A Test-Time Scaling Framework for Visual Auto-Regressive Generation

GoT: Unleashing Reasoning Capability of MLLM for Visual Generation and Editing

WorldSimBench: Towards Video Generation Models as World Simulators

Accelerating Auto-regressive Text-to-Image Generation with Training-free Speculative Jacobi Decoding

WorldSimBench: Towards Video Generation Models as World Simulators

Tailor3D: Customized 3D Assets Editing and Generation with Dual-Side Images

PUMA: Empowering Unified MLLM with Multi-granular Visual Generation

Benchmarking Ethics in Text-to-Image Models: A Holistic Dataset and Evaluator for Fairness, Toxicity, and Privacy