12 papers
Uncertainty DMD: Restoring Diversity in Few-Step Autoregressive Video Distillation
Zixuan Duan, Xunzhi Xiang, Yabo Chen +6
Few-step distillation improves the efficiency of autoregressive (AR) video generation, but often causes diversity collapse: under the same prompt, different noise samples tend to p…
Search-to-World: Evaluation of 3D World Delivery from User Request through Web Search
Zixiao Gu, Yabo Chen, Xunzhi Xiang +5
Agentic systems can interpret user requests, search the live web, and use external tools, but their ability to transform retrieved web content into a usable 3D world has not been s…
PhyEditBench: A Real-World Multi-Stage Benchmark for Physics-Aware Image Editing
Shengbin Guo, Shaokang He, Chaoyue Meng +4
While instruction-based image editing, enabled by multi-modal generative models, has advanced significantly, existing benchmarks lack a comprehensive evaluation of physics-based re…
VideoWeave: Unlocking Geometric Consistency in Video Generation via Joint Geometry-Video Modeling
Xunzhi Xiang, Zixuan Duan, Yabo Chen +8
Large-scale video diffusion models often fail to preserve 3D structure over time, causing geometric drift and implausible motion under viewpoint changes. Existing methods usually e…
Geometry-Aware Implicit Memory for Video World Models
Zhengxuan Wei, Xu Guo, Xinghui Li +8
Video world models aim to simulate controllable visual environments, but long-horizon rollouts depend on what the model remembers after observations leave its native context window…
Selective, Regularized, and Calibrated: Harnessing Vision Foundation Models for Cross-Domain Few-Shot Semantic Segmentation
Junyuan Ma, Xunzhi Xiang, Wenbin Li +2
Vision foundation models (VFMs) have achieved strong performance across various vision tasks. However, it still remains challenging to apply VFMs for cross-domain few-shot segmenta…
SymphoMotion: Joint Control of Camera Motion and Object Dynamics for Coherent Video Generation
Guiyu Zhang, Yabo Chen, Xunzhi Xiang +3
Controlling both camera motion and object dynamics is essential for coherent and expressive video generation, yet current methods typically handle only one motion type or rely on a…
Pathwise Test-Time Correction for Autoregressive Long Video Generation
Xunzhi Xiang, Zixuan Duan, Guiyu Zhang +7
Distilled autoregressive diffusion models facilitate real-time short video synthesis but suffer from severe error accumulation during long-sequence generation. While existing Test-…
Denoising Vision Transformer Autoencoder with Spectral Self-Regularization
Xunzhi Xiang, Xingye Tian, Guiyu Zhang +5
Variational autoencoders (VAEs) typically encode images into a compact latent space, reducing computational cost but introducing an optimization dilemma: a higher-dimensional laten…
Macro-from-Micro Planning for High-Quality and Parallelized Autoregressive Long Video Generation
Xunzhi Xiang, Yabo Chen, Guiyu Zhang +10
Current autoregressive diffusion models excel at video generation but are generally limited to short temporal durations. Our theoretical analysis indicates that the autoregressive…
Make It Efficient: Dynamic Sparse Attention for Autoregressive Image Generation
Xunzhi Xiang, Qi Fan
Autoregressive conditional image generation models have emerged as a dominant paradigm in text-to-image synthesis. These methods typically convert images into one-dimensional token…
Proteus-ID: ID-Consistent and Motion-Coherent Video Customization
Guiyu Zhang, Chen Shi, Zijian Jiang +4
Video identity customization seeks to synthesize realistic, temporally coherent videos of a specific subject, given a single reference image and a text prompt. This task presents t…