Parallel Rollout Approximation for Pixel-Space Autoregressive Image Generation
cs.CV, cs.AI
Submitted: 2026-06-26
Updated: 2026-11-07
Code: https://github.com/MangataX/PRA
Terminology
Sources
- PixelFlow: Pixel-Space Generative Models with Flow
- Adam: A Method for Stochastic Optimization
- Back to Basics: Let Denoising Generative Models Denoise
- Fractal Generative Models
- Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow
- Decoupled Weight Decay Regularization
- Unified Cross-Scale 3D Generation and Understanding via Autoregressive Modeling
- Autoregressive Speech Synthesis without Vector Quantization
- GLU Variants Improve Transformer
- Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
- Multimodal Latent Language Modeling with Next-Token Diffusion
- NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale
- MAGI-1: Autoregressive Video Generation at Scale
- JetFormer: An Autoregressive Generative Model of Raw Images and Text
- PixNerd: Pixel Neural Field Diffusion
- DDT: Decoupled Diffusion Transformer
- Vector-quantized Image Modeling with Improved VQGAN
- Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
- Diffusion Transformers with Representation Autoencoders
- FARMER: Flow AutoRegressive Transformer over Pixels
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs