Awesome-Autoregressive-Visual-Generation

This is a repo to track the latest autoregressive visual generation papers 🌟.

❗️❗️❗️In addition, I am also maintaining an open-source high-performance image tokenization framework XQ-GAN 🚀. This framework supports both state-of-the-art multi-scale residual quantization (VAR/Infinity) and regular vector quantizations (VQ-GAN variants).

Image Tokenizers

1. Neural Discrete Representation Learning Paper, NeurIPS 2017
2. Generating Diverse High-Fidelity Images with VQ-VAE-2 Paper, NeurIPS 2019
3. Taming Transformers for High-Resolution Image Synthesis Paper, CVPR 2021
4. Autoregressive Image Generation using Residual Quantization Paper, CVPR 2022
5. * BEIT V2: Masked Image Modeling with Vector-Quantized Visual Tokenizers (for understanding) Paper, Arxiv 2022
6. Vector-quantized Image Modeling with Improved VQGAN Paper, ICLR 2022
7. MoVQ: Modulating Quantized Vectors for High-Fidelity Image Generation Paper, NeurIPS 2022
8. * PeCo: Perceptual Codebook for BERT Pre-training of Vision Transformers (for understanding) Paper, AAAI 2023
9. * All in Tokens: Unifying Output Space of Visual Tasks via Soft Token (for understanding) Paper, CVPR 2023
10. Regularized Vector Quantization for Tokenized Image Synthesis Paper, CVPR 2023
11. Towards Accurate Image Coding: Improved Autoregressive Image Generation with Dynamic Vector Quantization Paper, CVPR 2023
12. Not all image regionsmatter: Masked vector quantization for autoregressive image generation Paper, CVPR 2023
13. Spae: Semantic pyramid autoencoder for multimodal generation with frozen llms Paper, NeurIPS 2023
14. HQ-VAE: Hierarchical Discrete Representation Learning with Variational Bayes Paper, TMLR 2024
15. Finite Scalar Quantization: VQ-VAE Made Simple Paper, ICLR 2024
16. Planting a seed of vision in large language model Paper, ICLR 2024
17. Language model beats diffusion–tokenizer is key to visual generation Paper, ICLR 2024
18. Rethinking the Objectives of Vector-Quantized Tokenizers for Image Synthesis Paper, CVPR 2024
19. Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction Paper, NeurIPS 2024
20. An Image is Worth 32 Tokens for Reconstruction and Generation Paper, NeurIPS 2024
21. Image Understanding Makes for A Good Tokenizer for Image Generation Paper, NeurIPS 2024
22. Scaling the Codebook Size of VQGAN to 100,000 with a Utilization Rate of 99% Paper, Arxiv 2024
23. Quantised Global Autoencoder: A Holistic Approach to Representing Visual Data Paper, Arxiv 2024
24. VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation Paper, ICLR 2025
25. OPEN-MAGVIT2: AN OPEN-SOURCE PROJECT TOWARD DEMOCRATIZING AUTO-REGRESSIVE VISUAL GENERATION Paper, Arxiv 2024
26. MaskBit: Embedding-free Image Generation via Bit Tokens Paper, TMLR 2024
27. ImageFolder: Autoregressive Image Generation with Folded Tokens 🚀 Paper, ICLR 2025
28. Cosmos Tokenizer: A suite of image and video neural tokenizers Website
29. Adaptive Length Image Tokenization via Recurrent Allocation Paper, Arxiv 2024
30. Stabilize the Latent Space for Image Autoregressive Modeling: A Unified Perspective Paper, Arxiv 2024
31. MUSE-VL: Modeling Unified VLM through Semantic Discrete Encoding Paper, Arxiv 2024
32. TokenFlow: Unified Image Tokenizer for Multimodal Understanding and Generation Paper, Arxiv 2024
33. XQ-GAN: An Open-source Image Tokenization Framework for Autoregressive Generation Paper, Arxiv 2024
34. Image and Video Tokenization with Binary Spherical Quantization Paper, Arxiv 2024
35. Taming Scalable Visual Tokenizer for Autoregressive Image Generation Paper, Arxiv 2024
36. Language-Guided Image Tokenization for Generation Paper, Arxiv 2024
37. Divot: Diffusion Powers Video Tokenizer for Comprehension and Generation Paper, Arxiv 2024
38. Scaling Image Tokenizers with Grouped Spherical Quantization Paper, Arxiv 2024
39. Taming Scalable Visual Tokenizer for Autoregressive Image Generation Paper, Arxiv 2024
40. Spectral Image Tokenizer Paper, Arxiv 2024
41. Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens Paper, Arxiv 2024
42. Learnings from Scaling Visual Tokenizers for Reconstruction and Generation Paper, Arxiv 2024
43. Language-Guided Image Tokenization for Generation Paper, Arxiv 2024
44. Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models Paper, Arxiv 2024
45. One-D-Piece: Image Tokenizer Meets Quality-Controllable Compression Paper, Arxiv 2024
46. QLIP: Text-Aligned Visual Tokenization Unifies Auto-Regressive Multimodal Understanding and Generation Paper, Arxiv 2025
47. FlexTok: Resampling Images into 1D Token Sequences of Flexible Length Paper, Arxiv 2025
48. UniTok: A Unified Tokenizer for Visual Generation and Understanding Paper, Arxiv 2025
49. NFIG: Autoregressive Image Generation with Next-Frequency Prediction Paper, Arxiv 2025
50. V2Flow: Unifying Visual Tokenization and Large Language Model Vocabularies for Autoregressive Image Generation Paper, Arxiv 2025
51. “Principal Components” Enable A New Language of Images Paper, Arxiv 2025
52. Flow to the Mode: Mode-Seeking Diffusion Autoencoders for State-of-the-Art Image Tokenization Paper, Arxiv 2025
53. Tokenize Image as a Set Paper, Arxiv 2025
54. D2C: Unlocking the Potential of Continuous Autoregressive Image Generation with Discrete Tokens, Paper, Arxiv 2025
55. Robust Latent Matters: Boosting Autoregressive Image Generation with Sampling Error Synthesis Paper, Arxiv 2025
56. Bridging Continuous and Discrete Tokens for Autoregressive Visual Generation Paper, Arxiv 2025
57. MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization Paper, Arxiv 2025
58. GigaTok: Scaling Visual Tokenizers to 3 Billion Parameters for Autoregressive Image Generation Paper, Arxiv 2025
59. MedITok: A Unified Tokenizer for Medical Image Synthesis and Interpretation Paper, Arxiv 2025

AutoRegressive Image Generation

1. Conditional image generation with pixelcnn decoders Paper, NeurIPS 2016
2. DiVAE : Photorealistic Images Synthesis with Denoising Diffusion Decoder Paper
3. Vector Quantized Diffusion Model for Text-to-Image Synthesis Paper
4. MaskGIT: Masked Generative Image Transformer Paper
5. BEIT: BERT Pre-Training of Image Transformers Paper
6. BEIT V2: Masked Image Modeling with Vector-Quantized Visual Tokenizers Paper
7. MAGE: MAsked Generative Encoder to Unify Representation Learning and Image Synthesis Paper
8. Sequential modeling enables scalable learning for large vision models Paper, Arxiv 2023
9. 4m: Massively multimodal masked modeling Paper, NeurIPS 2023
10. Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation Paper, Arxiv 2024
11. ControlVAR: Exploring Controllable Visual Autoregressive Modeling Paper, Arxiv 2024
12. Autoregressive Image Generation without Vector Quantization Paper, Arxiv 2024
13. MARS: Mixture of Auto-Regressive Models for Fine-grained Text-to-image Synthesis Paper, Arxiv 2024
14. ANOLE: AnOpen,Autoregressive, Native Large Multimodal Models for Interleaved Image-Text Generation Paper, Arxiv 2024
15. VAR-CLIP: Text-to-Image Generator with Visual Auto-Regressive Modeling Paper, Arxiv 24
16. Lumina-mGPT: Illuminate Flexible Photorealistic Text-to-Image Generation with Multimodal Generative Pretraining Paper, Arxiv 24
17. Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model Paper, Arxiv 2024
18. Scalable Autoregressive Image Generation with Mamba Paper, Arxiv 2024
19. SHOW-O: ONE SINGLE TRANSFORMER TO UNIFY MULTIMODAL UNDERSTANDING AND GENERATION Paper, ICLR 2025
20. DART: Denoising Autoregressive Transformer for Scalable Text-to-Image Generation Paper, Arxiv 2024
21. Fluid: Scaling Autoregressive Text-to-image Generative Models with Continuous Tokens Paper, Arxiv 2024
22. Customize Your Visual Autoregressive Recipe with Set Autoregressive Modeling Paper, Arxiv 2024
23. M-VAR: Decoupled Scale-wise Autoregressive Modeling for High-Quality Image Generation Paper, Arxiv 2024
24. MMAR:TowardsLossless Multi-Modal Auto-Regressive Probabilistic Modeling Paper, Arxiv 2024
25. Randomized Autoregressive Visual Generation Paper, Arxiv 2024
26. Elucidating the design space of language models for image generation Paper, Arxiv 2024
27. Toward Guidance-Free AR Visual Generation via Condition Contrastive Alignment Paper, ICLR 2025
28. CART: Compositional Auto-Regressive Transformer for Image Generation Paper, Arxiv 2024
29. CoDe: Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient Paper, Arxiv 2024
30. X-Prompt: Towards Universal In-Context Image Generation in Auto-Regressive Vision Language Foundation Models Paper, Arxiv 2024
31. JetFormer: An Autoregressive Generative Model of Raw Images and Text Paper, Arxiv 2024
32. Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis Paper, Arxiv 2024
33. Liquid: Language Models are Scalable Multi-modal Generators Paper, Arxiv 2024
34. RandAR: Decoder-only Autoregressive Visual Generation in Random Orders Paper, Arxiv 2024
35. FlowAR: Scale-wise Autoregressive Image Generation Meets Flow Matching Paper, Arxiv 2024
36. Parallelized Autoregressive Visual Generation Paper, Arxiv 2024
37. Next Patch Prediction for Autoregressive Visual Generation Paper, Arxiv 2024
38. MetaMorph: Multimodal Understanding and Generation via Instruction Tuning Paper, Arxiv 2024
39. VARGPT: Unified Understanding and Generation in a Visual Autoregressive Multimodal Large Language Model Paper, Arxiv 2024
40. Can We Generate Images with CoT? Let’s Verify and Reinforce Image Generation Step by Step Paper, Arxiv 2024
41. Generative Autoregressive Transformers for Model-Agnostic Federated MRI Reconstruction Paper, Arxiv 2025
42. Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling Paper, Arxiv 2025
43. VARGPT: Unified Understanding and Generation in a Visual Autoregressive Multimodal Large Language Model Paper, Arxiv 2025
44. Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation Paper, Arxiv 2025
45. Autoregressive Image Generation Guided by Chains of Thought Paper, Arxiv 2025
46. Fractal Generative Models Paper, Arxiv 2025
47. FlexVAR: Flexible Visual Autoregressive Modeling without Residual Prediction Paper, Arxiv 2025
48. Direction-Aware Diagonal Autoregressive Image Generation Paper, Arxiv 2025
49. DualToken: Towards Unifying Visual Understanding and Generation with Dual Visual Vocabularies Paper, Arxiv 2025
50. Unified Autoregressive Visual Generation and Understanding with Continuous Tokens Paper, Arxiv 2025
51. Autoregressive Image Generation with Randomized Parallel Decoding Paper, Arxiv 2025
52. Neighboring Autoregressive Modeling for Efficient Visual Generation Paper, Arxiv 2025
53. Beyond Words: Advancing Long-Text Image Generation via Multimodal Autoregressive Models Paper, Arxiv 2025
54. Harmonizing Visual Representations for Unified Multimodal Understanding and GenerationPaper, Arxiv 2025
55. FastVAR: Linear Visual Autoregressive Modeling via Cached Token Pruning Paper, Arxiv 2025
56. Training-Free Text-Guided Image Editing with Visual Autoregressive Model Paper, Arxiv 2025
57. Transfer between Modalities with MetaQueries Paper, Arxiv 2025
58. PixelFlow: Pixel-Space Generative Models with Flow, Paper, Arxiv 2025
59. Token-Shuffle: Towards High-Resolution Image Generation with Autoregressive Models Paper, Arxiv 2025
60. LazyMAR: Accelerating Masked Autoregressive Models via Feature Caching Paper, Arxiv 2025
61. Fast Autoregressive Models for Continuous Latent Generation Paper, Arxiv 2025
62. X-Fusion: Introducing New Modality to Frozen Large Language Models Paper, Arxiv 2025
63. T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT Paper, Arxiv 2025
64. Continuous Visual Autoregressive Generation via Score Maximization Paper, ICML 2025
65. DetailFlow: 1D Coarse-to-Fine Autoregressive Image Generation via Next-Detail Prediction Paper, Arxiv 2025
66. BLIP3-o: A Family of Fully Open Unified Multimodal Models—Architecture, Training and Dataset Paper, Arxiv 2025
67. DiSA: Diffusion Step Annealing in Autoregressive Image Generation Paper, Arixv 2025
68. StyleAR: Customizing Multimodal Autoregressive Model for Style-Aligned Text-to-Image Generation Paper, Arxiv 2025
69. TensorAR: Refinement is All You Need in Autoregressive Image Generation Paper, Arxiv 2025
70. D-AR: Diffusion via Autoregressive Models Paper, Arxiv 2025