# Bridging Continuous and Discrete Tokens for Autoregressive Visual Generation

arXiv: https://arxiv.org/abs/2503.16430 · HTML: https://mlsanity.com/paper/2503.16430/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- Builds on 47 of its 58 references that are in the map.
- First posted on arXiv 20 Mar 2025.

## Builds on (most cited first, 30 of 47)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Generating Diverse High-Fidelity Images with VQ-VAE-2](https://mlsanity.com/paper/1906.00446/) (2019)
- [MaskGIT: Masked Generative Image Transformer](https://mlsanity.com/paper/2202.04200/) (2022)
- [Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction](https://mlsanity.com/paper/2404.02905/) (2024)
- [Chameleon: Mixed-Modal Early-Fusion Foundation Models](https://mlsanity.com/paper/2405.09818/) (2024)
- [Pixel Recurrent Neural Networks](https://mlsanity.com/paper/1601.06759/) (2016)
- [Emu3: Next-Token Prediction is All You Need](https://mlsanity.com/paper/2409.18869/) (2024)
- [Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation](https://mlsanity.com/paper/2406.06525/) (2024)
- [Autoregressive Image Generation using Residual Quantization](https://mlsanity.com/paper/2203.01941/) (2022)
- [Vector-quantized Image Modeling with Improved VQGAN](https://mlsanity.com/paper/2110.04627/) (2021)
- [Autoregressive Image Generation without Vector Quantization](https://mlsanity.com/paper/2406.11838/) (2024)
- [Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation](https://mlsanity.com/paper/2310.05737/) (2023)
- [Scaling Laws for Autoregressive Generative Modeling](https://mlsanity.com/paper/2010.14701/) (2020)
- [All are Worth Words: A ViT Backbone for Diffusion Models](https://mlsanity.com/paper/2209.12152/) (2022)
- [VideoPoet: A Large Language Model for Zero-Shot Video Generation](https://mlsanity.com/paper/2312.14125/) (2023)
- [Phenaki: Variable Length Video Generation From Open Domain Textual Description](https://mlsanity.com/paper/2210.02399/) (2022)
- [Finite Scalar Quantization: VQ-VAE Made Simple](https://mlsanity.com/paper/2309.15505/) (2023)
- [PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modifications](https://mlsanity.com/paper/1701.05517/) (2017)
- [Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models](https://mlsanity.com/paper/2501.01423/) (2025)

## Built on by (most cited first, 29 of 29)

- [GoT-R1: Unleashing Reasoning Capability of MLLM for Visual Generation with Reinforcement Learning](https://mlsanity.com/paper/2505.17022/) (2025)
- [CoMo: Learning Continuous Latent Motion from Internet Videos for Scalable Robot Learning](https://mlsanity.com/paper/2505.17006/) (2025)
- [MotionGPT3: Human Motion as a Second Modality](https://mlsanity.com/paper/2506.24086/) (2025)
- [AToken: A Unified Tokenizer for Vision](https://mlsanity.com/paper/2509.14476/) (2025)
- [Kronos: A Foundation Model for the Language of Financial Markets](https://mlsanity.com/paper/2508.02739/) (2025)
- [TokBench: Evaluating Your Visual Tokenizer before Visual Generation](https://mlsanity.com/paper/2505.18142/) (2025)
- [FreqPolicy: Frequency Autoregressive Visuomotor Policy with Continuous Tokens](https://mlsanity.com/paper/2506.01583/) (2025)
- [Autoregressive Image Generation with Masked Bit Modeling](https://mlsanity.com/paper/2602.09024/) (2026)
- [Discrete Tokenization for Multimodal LLMs: A Comprehensive Survey](https://mlsanity.com/paper/2507.22920/) (2025)
- [LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model](https://mlsanity.com/paper/2604.02097/) (2026)
- [Quantize-then-Rectify: Efficient VQ-VAE Training](https://mlsanity.com/paper/2507.10547/) (2025)
- [Soft-Di[M]O: Improving One-Step Discrete Image Generation with Soft Embeddings](https://mlsanity.com/paper/2509.22925/) (2025)
- [PosterCopilot: Toward Layout Reasoning and Controllable Editing for Professional Graphic Design](https://mlsanity.com/paper/2512.04082/) (2025)
- [BitDance: Scaling Autoregressive Generative Models with Binary Tokens](https://mlsanity.com/paper/2602.14041/) (2026)
- [ClusterMark: Towards Robust Watermarking for Autoregressive Image Generators with Visual Token Clustering](https://mlsanity.com/paper/2508.06656/) (2025)
- [STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation](https://mlsanity.com/paper/2605.08029/) (2026)
- [Knowledge Distillation for Visual Autoregressive Models](https://mlsanity.com/paper/2606.06078/) (2026)
- [multivariateGPT: a decoder-only transformer for multivariate categorical and numeric data](https://mlsanity.com/paper/2505.21680/) (2025)
- [LSRS: Latent Scale Rejection Sampling for Visual Autoregressive Modeling](https://mlsanity.com/paper/2512.03796/) (2025)
- [Training-Free Vector Quantization via Gaussian VAEs](https://mlsanity.com/paper/2512.06609/) (2025)
- [VP-VAE: Rethinking Vector Quantization via Adaptive Vector Perturbation](https://mlsanity.com/paper/2602.17133/) (2026)
- [Cubic Discrete Diffusion: Discrete Visual Generation on High-Dimensional Representation Tokens](https://mlsanity.com/paper/2603.19232/) (2026)
- [On the Robustness of Watermarking for Autoregressive Image Generation](https://mlsanity.com/paper/2604.11720/) (2026)
- [DiverseVAR: Balancing Diversity and Quality of Next-Scale Visual Autoregressive Models](https://mlsanity.com/paper/2511.21415/) (2025)
- [High-Fidelity Text-to-Image Generation from Pre-Trained Vision-Language Models via Distribution-Conditioned Diffusion Decoding](https://mlsanity.com/paper/2603.13389/) (2026)
- [Continuous First, Discrete Later: VQ-VAEs Without Dimensional Collapse](https://mlsanity.com/paper/2605.06870/) (2026)
- [MergeTok: Unified Continuous and Discrete Visual Tokenization via Token Merging](https://mlsanity.com/paper/2605.30904/) (2026)
- [NSVQ: Mitigating Codebook Collapse by Stabilizing Encoder Drift in Vector Quantization](https://mlsanity.com/paper/2606.11363/) (2026)
- [Attribute Token Arithmetic: Disentangled and Continuous Semantic Control for Visual Autoregressive Models](https://mlsanity.com/paper/2608.28082/) (2026)

## Abstract (arXiv)

Autoregressive visual generation models typically rely on tokenizers to compress images into tokens that can be predicted sequentially. A fundamental dilemma exists in token representation: discrete tokens enable straightforward modeling with standard cross-entropy loss, but suffer from information loss and tokenizer training instability; continuous tokens better preserve visual details, but require complex distribution modeling, complicating the generation pipeline. In this paper, we propose TokenBridge, which bridges this gap by maintaining the strong representation capacity of continuous tokens while preserving the modeling simplicity of discrete tokens. To achieve this, we decouple discretization from the tokenizer training process through post-training quantization that directly obtains discrete tokens from continuous representations. Specifically, we introduce a dimension-wise quantization strategy that independently discretizes each feature dimension, paired with a lightweight autoregressive prediction mechanism that efficiently model the resulting large token space. Extensive experiments show that our approach achieves reconstruction and generation quality on par with continuous methods while using standard categorical prediction. This work demonstrates that bridging discrete and continuous paradigms can effectively harness the strengths of both approaches, providing a promising direction for high-quality visual generation with simple autoregressive modeling. Project page: https://yuqingwang1029.github.io/TokenBridge.