Showing 2026Show all
2 papers · 1 filter
cs.CV2026
UniGP: Taming Diffusion Transformer for Prior-Preserved Unified Generation and Perception
Qin Guo, Hao Luo, Dongxu Yue +4
Recent advances in diffusion models have shown impressive performance in controllable image generation and dense prediction tasks. However, existing approaches typically treat diff…
cs.CV2026
Train Short, Inference Long: Training-free Horizon Extension for Autoregressive Video Generation
Jia Li, Xiaomeng Fu, Xurui Peng +7
Autoregressive video diffusion models have emerged as a scalable paradigm for long video generation. However, they often suffer from severe extrapolation failure, where rapid error…