3 papers
cs.CV2025
X-Omni: Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Again
Zigang Geng, Yibing Wang, Yeyao Ma +10
Numerous efforts have been made to extend the ``next token prediction'' paradigm to visual contents, aiming to create a unified approach for both image generation and understanding…
cs.CV2025
Optimal Stepsize for Diffusion Sampling
Jianning Pei, Han Hu, Shuyang Gu
Diffusion models achieve remarkable generation quality but suffer from computational intensive sampling due to suboptimal step discretization. While existing works focus on optimiz…
cs.CV2025
Tokenize Image as a Set
Zigang Geng, Mengde Xu, Han Hu +1
This paper proposes a fundamentally new paradigm for image generation through set-based tokenization and distribution modeling. Unlike conventional methods that serialize images in…