Xin Tao
- Papers
- 18
Cite
Notes
Only stored in your browser.
Authored papers
18A Mechanistic View on Video Generation as World Models: State and Dynamics
arXiv 2026
Stable Velocity: A Variance Perspective on Flow Matching
arXiv 2026
Training-Free Efficient Video Generation via Dynamic Token Carving
arXiv 2025
RectifiedHR: Enable Efficient High-Resolution Image Generation via Energy Rectification
arXiv 2025
PhysMaster: Mastering Physical Representation for Video Generation via Reinforcement Learning
arXiv 2025
UnityVideo: Unified Multi-Modal Multi-Task Learning for Enhancing World-Aware Video Generation
arXiv 2025
StereoPilot: Learning Unified and Efficient Stereo Conversion via Generative Priors
arXiv 2025
MemFlow: Flowing Adaptive Memory for Consistent and Efficient Long Video Narratives
arXiv 2025
DiffMoE: Dynamic Token Selection for Scalable Diffusion Transformers
arXiv 2025
Video-as-Answer: Predict and Generate Next Video Event with Joint-GRPO
arXiv 2025
Alchemist: Unlocking Efficiency in Text-to-Image Model Training via Meta-Gradient Data Selection
arXiv 2025
Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention
arXiv 2025
VMoBA: Mixture-of-Block Attention for Video Diffusion Models
arXiv 2025
Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play?
arXiv 2025
VideoTetris: Towards Compositional Text-to-Video Generation
arXiv 2024
DVIS++: Improved Decoupled Framework for Universal Video Segmentation
arXiv 2023
VCNet: A Robust Approach to Blind Image Inpainting
ECCV 2020 8
Image Inpainting via Generative Multi-column Convolutional Neural Networks
image-inpainting-via-generative-multi-column-1
Affiliations
Frequent co-authors
10from 18 papers