mlsanityData to 7 Oct 2026

Channel-wise Vector Quantization

arXiv 2605.26089 ↗First posted 25 May 2026cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

We present Channel-wise Vector Quantization (CVQ), a novel image tokenization paradigm that replaces patch-wise tokens with channel-wise tokens. Unlike conventional vector quantization, which assigns a discrete token to each patch feature vector, CVQ quantizes each channel of the feature map. This formulation represents an image as discrete levels of visual details, rather than as a grid of spatial patches. Based on CVQ, we introduce a new visual autoregressive framework with "next-channel prediction". Instead of rendering images patch by patch in raster order, our Channel-wise Autoregressive (CAR) model predicts image channels sequentially, producing progressively enriched visual details. Specifically, it first sketches global structure and then refines fine-grained attributes, akin to a human artist's workflow. Empirically, we show that: (1) CVQ achieves 100% codebook utilization with a 16K+ codebook size without any bells and whistles, and substantially improves reconstruction quality over conventional VQ; and (2) CAR attains a DPG score of 86.7 and a GenEval score of 0.79, demonstrating strong effectiveness for text-to-image generation.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 55 of its 60 references that are in the map. First posted on arXiv 25 May 2026.

Builds on

55 of 60 references.

References · 55

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  4. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  5. 2025Qwen3 Technical Reportcited by 8,195
  6. 2017Neural Discrete Representation Learningcited by 4,650
  7. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  8. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  9. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  10. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  11. 2021Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Conceptscited by 1,209
  12. 2022MaskGIT: Masked Generative Image Transformercited by 1,082
38 more
  1. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,005
  2. 2024Chameleon: Mixed-Modal Early-Fusion Foundation Modelscited by 904
  3. 2024Emu3: Next-Token Prediction is All You Needcited by 761
  4. 2024Autoregressive Model Beats Diffusion: Llama for Scalable Image Generationcited by 760
  5. 2024Show-o: One Single Transformer to Unify Multimodal Understanding and Generationcited by 734
  6. 2022Autoregressive Image Generation using Residual Quantizationcited by 706
  7. 2021Vector-quantized Image Modeling with Improved VQGANcited by 656
  8. 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 633
  9. 2024Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generationcited by 447
  10. 2023Finite Scalar Quantization: VQ-VAE Made Simplecited by 425
  11. 2025BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Datasetcited by 363
  12. 2024SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformerscited by 331
  13. 2024Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesiscited by 297
  14. 2024An Image is Worth 32 Tokens for Reconstruction and Generationcited by 295
  15. 2024VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generationcited by 290
  16. 2024TokenFlow: Unified Image Tokenizer for Multimodal Understanding and Generationcited by 200
  17. 2025UniTok: A Unified Tokenizer for Visual Generation and Understandingcited by 148
  18. 2024Lumina-Next: Making Lumina-T2X Stronger and Faster with Next-DiTcited by 130
  19. 2024HART: Efficient Visual Generation with Hybrid Autoregressive Transformercited by 128
  20. 2022MoVQ: Modulating Quantized Vectors for High-Fidelity Image Generationcited by 125
  21. 2024Randomized Autoregressive Visual Generationcited by 125
  22. 2025FlexTok: Resampling Images into 1D Token Sequences of Flexible Lengthcited by 105
  23. 2024RandAR: Decoder-only Autoregressive Visual Generation in Random Orderscited by 82
  24. 2014Learning Ordered Representations with Nested Dropoutcited by 81
  25. 2024Addressing Representation Collapse in Vector Quantized Models with One Linear Layercited by 77
  26. 2024Liquid: Language Models are Scalable and Unified Multi-modal Generatorscited by 62
  27. 2023Towards Accurate Image Coding: Improved Autoregressive Image Generation with Dynamic Vector Quantizationcited by 59
  28. 2025NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scalecited by 55
  29. 2024Restructuring Vector Quantization with the Rotation Trickcited by 49
  30. 2025Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokenscited by 48
  31. 2024MUSE-VL: Modeling Unified VLM through Semantic Discrete Encodingcited by 44
  32. 2024Scalable Image Tokenization with Index Backpropagation Quantizationcited by 44
  33. 2025InfinityStar: Unified Spacetime AutoRegressive Modeling for Visual Generationcited by 43
  34. 2025DualToken: Towards Unifying Visual Understanding and Generation with Dual Visual Vocabulariescited by 42
  35. 2024Switti: Designing Scale-Wise Transformers for Text-to-Image Synthesiscited by 33
  36. 2025FlowTok: Flowing Seamlessly Across Text and Image Tokenscited by 31
  37. 2025"Principal Components" Enable A New Language of Imagescited by 25
  38. 2025VARGPT-v1.1: Improve Visual Autoregressive Large Unified Model via Iterative Instruction Tuning and Reinforcement Learningcited by 21

50 most cited of 55 All 55 (JSON)

Built on by

Not cited yet.