# Rethinking the Objectives of Vector-Quantized Tokenizers for Image Synthesis

arXiv: https://arxiv.org/abs/2212.03185 · HTML: https://mlsanity.com/paper/2212.03185/

## Key facts

- Cited by 31 papers in the map (data through 14 Oct 2026).
- Builds on 63 of its 70 references that are in the map.
- First posted on arXiv 6 Dec 2022.

## Builds on (most cited first, 30 of 63)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (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)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://mlsanity.com/paper/1603.08155/) (2016)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [Improved Baselines with Momentum Contrastive Learning](https://mlsanity.com/paper/2003.04297/) (2020)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [Structured Denoising Diffusion Models in Discrete State-Spaces](https://mlsanity.com/paper/2107.03006/) (2021)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)

## Built on by (most cited first, 30 of 31)

- [Show-o: One Single Transformer to Unify Multimodal Understanding and Generation](https://mlsanity.com/paper/2408.12528/) (2024)
- [Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models](https://mlsanity.com/paper/2501.01423/) (2025)
- [MDTv2: Masked Diffusion Transformer is a Strong Image Synthesizer](https://mlsanity.com/paper/2303.14389/) (2023)
- [MMaDA: Multimodal Large Diffusion Language Models](https://mlsanity.com/paper/2505.15809/) (2025)
- [Making LLaMA SEE and Draw with SEED Tokenizer](https://mlsanity.com/paper/2310.01218/) (2023)
- [Long-Context Autoregressive Video Modeling with Next-Frame Prediction](https://mlsanity.com/paper/2503.19325/) (2025)
- [MaskBit: Embedding-free Image Generation via Bit Tokens](https://mlsanity.com/paper/2409.16211/) (2024)
- [Masked Autoencoders Are Effective Tokenizers for Diffusion Models](https://mlsanity.com/paper/2502.03444/) (2025)
- [ImageFolder: Autoregressive Image Generation with Folded Tokens](https://mlsanity.com/paper/2410.01756/) (2024)
- [All-in-one simulation-based inference](https://mlsanity.com/paper/2404.09636/) (2024)
- [SoftVQ-VAE: Efficient 1-Dimensional Continuous Tokenizer](https://mlsanity.com/paper/2412.10958/) (2024)
- [Online Clustered Codebook](https://mlsanity.com/paper/2307.15139/) (2023)
- [LARP: Tokenizing Videos with a Learned Autoregressive Generative Prior](https://mlsanity.com/paper/2410.21264/) (2024)
- [Image Understanding Makes for A Good Tokenizer for Image Generation](https://mlsanity.com/paper/2411.04406/) (2024)
- [When Worse is Better: Navigating the compression-generation tradeoff in visual tokenization](https://mlsanity.com/paper/2412.16326/) (2024)
- [Multimodal Medical Code Tokenizer](https://mlsanity.com/paper/2502.04397/) (2025)
- [Representation Alignment for Just Image Transformers is not Easier than You Think](https://mlsanity.com/paper/2603.14366/) (2026)
- [XQ-GAN: An Open-source Image Tokenization Framework for Autoregressive Generation](https://mlsanity.com/paper/2412.01762/) (2024)
- [Factorized Visual Tokenization and Generation](https://mlsanity.com/paper/2411.16681/) (2024)
- [Greedy-DiM: Greedy Algorithms for Unreasonably Effective Face Morphs](https://mlsanity.com/paper/2404.06025/) (2024)
- [LG-VQ: Language-Guided Codebook Learning](https://mlsanity.com/paper/2405.14206/) (2024)
- [Discrete Tokenization for Multimodal LLMs: A Comprehensive Survey](https://mlsanity.com/paper/2507.22920/) (2025)
- [DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space](https://mlsanity.com/paper/2509.25180/) (2025)
- [Rethinking Discrete Tokens: Treating Them as Conditions for Continuous Autoregressive Image Synthesis](https://mlsanity.com/paper/2507.01756/) (2025)
- [UniMoD: Efficient Unified Multimodal Transformers with Mixture-of-Depths](https://mlsanity.com/paper/2502.06474/) (2025)
- [Towards Improved Text-Aligned Codebook Learning: Multi-Hierarchical Codebook-Text Alignment with Long Text](https://mlsanity.com/paper/2503.01261/) (2025)
- [BiomechGPT: Extending Motion-Language Models to Clinical Motion Understanding](https://mlsanity.com/paper/2505.18465/) (2025)
- [LooC: Effective Low-Dimensional Codebook for Compositional Vector Quantization](https://mlsanity.com/paper/2601.00222/) (2026)
- [Is Hierarchical Quantization Essential for Optimal Reconstruction?](https://mlsanity.com/paper/2601.22244/) (2026)
- [RAC: Rectified Flow Auto Coder](https://mlsanity.com/paper/2603.05925/) (2026)

## Abstract (arXiv)

Vector-Quantized (VQ-based) generative models usually consist of two basic components, i.e., VQ tokenizers and generative transformers. Prior research focuses on improving the reconstruction fidelity of VQ tokenizers but rarely examines how the improvement in reconstruction affects the generation ability of generative transformers. In this paper, we surprisingly find that improving the reconstruction fidelity of VQ tokenizers does not necessarily improve the generation. Instead, learning to compress semantic features within VQ tokenizers significantly improves generative transformers' ability to capture textures and structures. We thus highlight two competing objectives of VQ tokenizers for image synthesis: semantic compression and details preservation. Different from previous work that only pursues better details preservation, we propose Semantic-Quantized GAN (SeQ-GAN) with two learning phases to balance the two objectives. In the first phase, we propose a semantic-enhanced perceptual loss for better semantic compression. In the second phase, we fix the encoder and codebook, but enhance and finetune the decoder to achieve better details preservation. The proposed SeQ-GAN greatly improves VQ-based generative models and surpasses the GAN and Diffusion Models on both unconditional and conditional image generation. Our SeQ-GAN (364M) achieves Frechet Inception Distance (FID) of 6.25 and Inception Score (IS) of 140.9 on 256x256 ImageNet generation, a remarkable improvement over VIT-VQGAN (714M), which obtains 11.2 FID and 97.2 IS.