STAR. STAR: Scale-wise Text-to-image generation via Auto-Regressive representations
150MPI. Code for paper "Masked Pre-training Enables Universal Zero-shot Denoiser" [NeurIPS 2024].
35STAGE. A stable & generalizable GRPO method for AR image generation
33ARsample. Code for paper "Towards Better & Faster Autoregressive Image Generation: From the Perspective of Entropy" [NeurIPS 2025] .
18