# Parallel Multiscale Autoregressive Density Estimation

arXiv: https://arxiv.org/abs/1703.03664 · HTML: https://mlsanity.com/paper/1703.03664/

## Key facts

- Cited by 71 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Mar 2017.

## Built on by (most cited first, 30 of 71)

- [Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction](https://mlsanity.com/paper/2404.02905/) (2024)
- [Generating Images with Sparse Representations](https://mlsanity.com/paper/2103.03841/) (2021)
- [Cross-Modal Contrastive Learning for Text-to-Image Generation](https://mlsanity.com/paper/2101.04702/) (2021)
- [Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models](https://mlsanity.com/paper/2103.04922/) (2021)
- [Autoregressive Diffusion Models](https://mlsanity.com/paper/2110.02037/) (2021)
- [Generative Modelling With Inverse Heat Dissipation](https://mlsanity.com/paper/2206.13397/) (2022)
- [Neural Video Compression with Diverse Contexts](https://mlsanity.com/paper/2302.14402/) (2023)
- [Reinforcement Learning with Action-Free Pre-Training from Videos](https://mlsanity.com/paper/2203.13880/) (2022)
- [FitVid: Overfitting in Pixel-Level Video Prediction](https://mlsanity.com/paper/2106.13195/) (2021)
- [Unified Multimodal Understanding and Generation Models: Advances, Challenges, and Opportunities](https://mlsanity.com/paper/2505.02567/) (2025)
- [PixelSynth: Generating a 3D-Consistent Experience from a Single Image](https://mlsanity.com/paper/2108.05892/) (2021)
- [ARTboldsymbol·V: Auto-Regressive Text-to-Video Generation with Diffusion Models](https://mlsanity.com/paper/2311.18834/) (2023)
- [3D Semantic Scene Completion: a Survey](https://mlsanity.com/paper/2103.07466/) (2021)
- [Learning to Compose Visual Relations](https://mlsanity.com/paper/2111.09297/) (2021)
- [Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes](https://mlsanity.com/paper/2111.12701/) (2021)
- [Combiner: Full Attention Transformer with Sparse Computation Cost](https://mlsanity.com/paper/2107.05768/) (2021)
- [Predicting Video with VQVAE](https://mlsanity.com/paper/2103.01950/) (2021)
- [Look Outside the Room: Synthesizing A Consistent Long-Term 3D Scene Video from A Single Image](https://mlsanity.com/paper/2203.09457/) (2022)
- [Adversarial Text-to-Image Synthesis: A Review](https://mlsanity.com/paper/2101.09983/) (2021)
- [Zero-Shot Detection of AI-Generated Images](https://mlsanity.com/paper/2409.15875/) (2024)
- [DAE-GAN: Dynamic Aspect-aware GAN for Text-to-Image Synthesis](https://mlsanity.com/paper/2108.12141/) (2021)
- [Autoregressive Models in Vision: A Survey](https://mlsanity.com/paper/2411.05902/) (2024)
- [A Survey on Graph Diffusion Models: Generative AI in Science for Molecule, Protein and Material](https://mlsanity.com/paper/2304.01565/) (2023)
- [Image Compression with Product Quantized Masked Image Modeling](https://mlsanity.com/paper/2212.07372/) (2022)
- [Deep Lossy Plus Residual Coding for Lossless and Near-lossless Image Compression](https://mlsanity.com/paper/2209.04847/) (2022)
- [HARP: Autoregressive Latent Video Prediction with High-Fidelity Image Generator](https://mlsanity.com/paper/2209.07143/) (2022)
- [Vector Quantized Wasserstein Auto-Encoder](https://mlsanity.com/paper/2302.05917/) (2023)
- [Multiscale deep context modeling for lossless point cloud geometry compression](https://mlsanity.com/paper/2104.09859/) (2021)
- [RestoreVAR: Visual Autoregressive Generation for All-in-One Image Restoration](https://mlsanity.com/paper/2505.18047/) (2025)
- [Fine-Tuning Visual Autoregressive Models for Subject-Driven Generation](https://mlsanity.com/paper/2504.02612/) (2025)

## Abstract (arXiv)

PixelCNN achieves state-of-the-art results in density estimation for natural images. Although training is fast, inference is costly, requiring one network evaluation per pixel; O(N) for N pixels. This can be sped up by caching activations, but still involves generating each pixel sequentially. In this work, we propose a parallelized PixelCNN that allows more efficient inference by modeling certain pixel groups as conditionally independent. Our new PixelCNN model achieves competitive density estimation and orders of magnitude speedup - O(log N) sampling instead of O(N) - enabling the practical generation of 512x512 images. We evaluate the model on class-conditional image generation, text-to-image synthesis, and action-conditional video generation, showing that our model achieves the best results among non-pixel-autoregressive density models that allow efficient sampling.