mlsanityData to 14 Oct 2026

Parallelized Autoregressive Visual Generation

arXiv 2412.15119 ↗First posted 19 Dec 2024cs.CVcited by 60

Not on the map. Why?

Abstract (arXiv)

Autoregressive models have emerged as a powerful approach for visual generation but suffer from slow inference speed due to their sequential token-by-token prediction process. In this paper, we propose a simple yet effective approach for parallelized autoregressive visual generation that improves generation efficiency while preserving the advantages of autoregressive modeling. Our key insight is that parallel generation depends on visual token dependencies-tokens with weak dependencies can be generated in parallel, while strongly dependent adjacent tokens are difficult to generate together, as their independent sampling may lead to inconsistencies. Based on this observation, we develop a parallel generation strategy that generates distant tokens with weak dependencies in parallel while maintaining sequential generation for strongly dependent local tokens. Our approach can be seamlessly integrated into standard autoregressive models without modifying the architecture or tokenizer. Experiments on ImageNet and UCF-101 demonstrate that our method achieves a 3.6x speedup with comparable quality and up to 9.5x speedup with minimal quality degradation across both image and video generation tasks. We hope this work will inspire future research in efficient visual generation and unified autoregressive modeling. Project page: https://yuqingwang1029.github.io/PAR-project.

Key facts. Cited by 60 papers in the map (data through 14 Oct 2026). Builds on 57 of its 74 references that are in the map. First posted on arXiv 19 Dec 2024.

Builds on

57 of 74 references.

References · 57

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  5. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  6. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  7. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  8. 2022Scalable Diffusion Models with Transformerscited by 6,054
  9. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  10. 2022Classifier-Free Diffusion Guidancecited by 5,847
  11. 2017Neural Discrete Representation Learningcited by 4,712
  12. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
38 more
  1. 2021Zero-Shot Text-to-Image Generationcited by 4,288
  2. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
  3. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  4. 2022Make-A-Video: Text-to-Video Generation without Text-Video Datacited by 1,742
  5. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
  6. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,295
  7. 2018Towards Accurate Generative Models of Video: A New Metric & Challengescited by 1,234
  8. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,163
  9. 2022MaskGIT: Masked Generative Image Transformercited by 1,099
  10. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,023
  11. 2023Accelerating Large Language Model Decoding with Speculative Samplingcited by 1,006
  12. 2022CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformerscited by 985
  13. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 944
  14. 2016Deep Variational Information Bottleneckcited by 930
  15. 2024Chameleon: Mixed-Modal Early-Fusion Foundation Modelscited by 927
  16. 2016Conditional Image Generation with PixelCNN Decoderscited by 862
  17. 2016Pixel Recurrent Neural Networkscited by 816
  18. 2015Deep Learning and the Information Bottleneck Principlecited by 802
  19. 2024Emu3: Next-Token Prediction is All You Needcited by 781
  20. 2024Autoregressive Model Beats Diffusion: Llama for Scalable Image Generationcited by 777
  21. 2018Image Transformercited by 772
  22. 2022Autoregressive Image Generation using Residual Quantizationcited by 722
  23. 2021VideoGPT: Video Generation using VQ-VAE and Transformerscited by 658
  24. 2021Vector-quantized Image Modeling with Improved VQGANcited by 658
  25. 2024Autoregressive Image Generation without Vector Quantizationcited by 651
  26. 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 640
  27. 2024EAGLE: Speculative Sampling Requires Rethinking Feature Uncertaintycited by 578
  28. 2020Scaling Laws for Autoregressive Generative Modelingcited by 522
  29. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 492
  30. 2023Scaling up GANs for Text-to-Image Synthesiscited by 492
  31. 2023VideoPoet: A Large Language Model for Zero-Shot Video Generationcited by 492
  32. 2017PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modificationscited by 417
  33. 2022MAGVIT: Masked Generative Video Transformercited by 346
  34. 2023VideoFusion: Decomposed Diffusion Models for High-Quality Video Generationcited by 331
  35. 2018Blockwise Parallel Decoding for Deep Autoregressive Modelscited by 323
  36. 2022Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformercited by 244
  37. 2024Lumina-mGPT: Illuminate Flexible Photorealistic Text-to-Image Generation with Multimodal Generative Pretrainingcited by 150
  38. 2024OmniTokenizer: A Joint Image-Video Tokenizer for Visual Generationcited by 118

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 30; 2026: 30.

60 papers.

Sort:

By area:

  1. 2025Autoregressive Adversarial Post-Training for Real-Time Interactive Video Generationcited by 90
  2. 2025Unified Multimodal Understanding and Generation Models: Advances, Challenges, and Opportunitiescited by 85
  3. 2025Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generationcited by 71
  4. 2025GoT-R1: Unleashing Reasoning Capability of MLLM for Visual Generation with Reinforcement Learningcited by 52
  5. 2025Neighboring Autoregressive Modeling for Efficient Visual Generationcited by 29
  6. 2025Bridging Continuous and Discrete Tokens for Autoregressive Visual Generationcited by 29
  7. 2025Frequency Autoregressive Image Generation with Continuous Tokenscited by 27
  8. 2025Next Block Prediction: Video Generation via Semi-Autoregressive Modelingcited by 26
  9. 2025Macro-from-Micro Planning for High-Quality and Parallelized Autoregressive Long Video Generationcited by 21
  10. 2025Autoregressive Image Generation with Randomized Parallel Decodingcited by 20
  11. 2025DetailFlow: 1D Coarse-to-Fine Autoregressive Image Generation via Next-Detail Predictioncited by 18
  12. 2025Grouped Speculative Decoding for Autoregressive Image Generationcited by 17
38 more
  1. 2025DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizercited by 13
  2. 2025MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantizationcited by 12
  3. 2025Image Tokenizer Needs Post-Trainingcited by 12
  4. 2025Towards Better & Faster Autoregressive Image Generation: From the Perspective of Entropycited by 12
  5. 2025LazyMAR: Accelerating Masked Autoregressive Models via Feature Cachingcited by 11
  6. 2025UniFork: Exploring Modality Alignment for Unified Multimodal Understanding and Generationcited by 11
  7. 2025Robust Latent Matters: Boosting Image Generation with Sampling Error Synthesiscited by 10
  8. 2025FastCar: Cache Attentive Replay for Fast Auto-Regressive Video Generation on the Edgecited by 10
  9. 2025Locality-aware Parallel Decoding for Efficient Autoregressive Image Generationcited by 10
  10. 2026Autoregressive Image Generation with Masked Bit Modelingcited by 9
  11. 2025Understand Before You Generate: Self-Guided Training for Autoregressive Image Generationcited by 7
  12. 2025Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generationcited by 7
  13. 2026VideoAR: Autoregressive Video Generation via Next-Frame & Scale Predictioncited by 7
  14. 2026Evolution of Video Generative Foundationscited by 7
  15. 2025DiCo: Revitalizing ConvNets for Scalable and Efficient Diffusion Modelingcited by 6
  16. 2025Rethinking Discrete Tokens: Treating Them as Conditions for Continuous Autoregressive Image Synthesiscited by 6
  17. 2026ResTok: Learning Hierarchical Residuals in 1D Visual Tokenizers for Autoregressive Image Generationcited by 5
  18. 2025Autoregressive Image Generation with Linear Complexity: A Spatial-Aware Decay Perspectivecited by 4
  19. 2026PTQ4ARVG: Post-Training Quantization for AutoRegressive Visual Generation Modelscited by 4
  20. 2026SparVAR: Exploring Sparsity in Visual AutoRegressive Modeling for Training-Free Accelerationcited by 4
  21. 2026BitDance: Scaling Autoregressive Generative Models with Binary Tokenscited by 4
  22. 2026Adaptive 1D Video Diffusion Autoencodercited by 3
  23. 2025SoftCFG: Uncertainty-guided Stable Guidance for Visual Autoregressive Modelcited by 2
  24. 2025Autoregressive Image Generation Needs Only a Few Lines of Cached Tokenscited by 2
  25. 2025Fast-ARDiff: An Entropy-informed Acceleration Framework for Continuous Space Autoregressive Generationcited by 2
  26. 2026Mirai: Autoregressive Visual Generation Needs Foresightcited by 2
  27. 2026Parallel Jacobi Decoding for Fast Autoregressive Image Generationcited by 2
  28. 2026SSD: Spatially Speculative Decoding Accelerates Autoregressive Image Generationcited by 2
  29. 2026Beyond Patches: Global-aware Autoregressive Model for Multimodal Few-Shot Font Generationcited by 1
  30. 2026SJD-PV: Speculative Jacobi Decoding with Phrase Verification for Autoregressive Image Generationcited by 1
  31. 2026Cubic Discrete Diffusion: Discrete Visual Generation on High-Dimensional Representation Tokenscited by 1
  32. 2026SJD-VP: Speculative Jacobi Decoding with Verification Prediction for Autoregressive Image Generationcited by 1
  33. 2025ARSS: Taming Decoder-only Autoregressive Visual Generation for View Synthesis From Single Viewcited by 0
  34. 2026Progressive Checkerboards for Autoregressive Multiscale Image Generationcited by 0
  35. 2026Revisiting Autoregressive Models for Generative Image Classificationcited by 0
  36. 2026Drift-AR: Single-Step Visual Autoregressive Generation via Anti-Symmetric Driftingcited by 0
  37. 2026Autoregressive Visual Generation Needs a Prologuecited by 0
  38. 2026Taming the Entropy Cliff: Variable Codebook Size Quantization for Autoregressive Visual Generationcited by 0

50 most cited of 60 All 60 (JSON)