Xihui Liu

46

Papers

1,260

Total Citations

Papers (46)

WorldSimBench: Towards Video Generation Models as World Simulators

LLaVA-3D: A Simple yet Effective Pathway to Empowering LMMs with 3D Capabilities

Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training

GameFactory: Creating New Games with Generative Interactive Videos

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

T2ISafety: Benchmark for Assessing Fairness, Toxicity, and Privacy in Image Generation

GigaTok: Scaling Visual Tokenizers to 3 Billion Parameters for Autoregressive Image Generation

DreamComposer: Controllable 3D Object Generation via Multi-View Conditions

PUMA: Empowering Unified MLLM with Multi-granular Visual Generation

Bridging Continuous and Discrete Tokens for Autoregressive Visual Generation

RoboFactory: Exploring Embodied Agent Collaboration with Compositional Constraints

MBQ: Modality-Balanced Quantization for Large Vision-Language Models

OST-Bench: Evaluating the Capabilities of MLLMs in Online Spatio-temporal Scene Understanding

Object Detection in Videos With Tubelet Proposal Networks

Improving Referring Expression Grounding With Cross-Modal Attention-Guided Erasing

HMAR: Efficient Hierarchical Masked Auto-Regressive Image Generation

Masked Scene Contrast: A Scalable Framework for Unsupervised 3D Representation Learning

GLeaD: Improving GANs With a Generator-Leading Task

RIFormer: Keep Your Vision Backbone Effective but Removing Token Mixer

Back to the Source: Diffusion-Driven Adaptation To Test-Time Corruption

Learning Transferable Spatiotemporal Representations From Natural Script Knowledge

HydraPlus-Net: Attentive Deep Features for Pedestrian Analysis

Orientation Invariant Feature Embedding and Spatial Temporal Regularization for Vehicle Re-Identification

CAMP: Cross-Modal Adaptive Message Passing for Text-Image Retrieval

DDP: Diffusion Model for Dense Visual Prediction

Open-Edit: Open-Domain Image Manipulation with Open-Vocabulary Instructions

MILES: Visual BERT Pre-training with Injected Language Semantics for Video-Text Retrieval

Bridging Video-Text Retrieval With Multiple Choice Questions

Parallelized Autoregressive Visual Generation

MIDI: Multi-Instance Diffusion for Single Image to 3D Scene Generation

T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generation

Moto: Latent Motion Token as the Bridging Language for Learning Robot Manipulation from Videos

LiT: Delving into a Simple Linear Diffusion Transformer for Image Generation

V2PE: Improving Multimodal Long-Context Capability of Vision-Language Models with Variable Visual Position Encoding

DreamCube: RGB-D Panorama Generation via Multi-plane Synchronization

EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI

Point Transformer V3: Simpler Faster Stronger

HumanGaussian: Text-Driven 3D Human Generation with Gaussian Splatting

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

FiT: Flexible Vision Transformer for Diffusion Model

Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis

Point Transformer V2: Grouped Vector Attention and Partition-based Pooling

Seeing is not always believing: Benchmarking Human and Model Perception of AI-Generated Images

CorresNeRF: Image Correspondence Priors for Neural Radiance Fields

OV-PARTS: Towards Open-Vocabulary Part Segmentation

T2I-CompBench: A Comprehensive Benchmark for Open-world Compositional Text-to-image Generation