mlsanityData to 7 Oct 2026

Exploration into Translation-Equivariant Image Quantization

arXiv 2112.00384 ↗First posted 1 Dec 2021cs.CV cs.CL cs.LGcited by 10

Not on the map. Why?

Abstract (arXiv)

This is an exploratory study that discovers the current image quantization (vector quantization) do not satisfy translation equivariance in the quantized space due to aliasing. Instead of focusing on anti-aliasing, we propose a simple yet effective way to achieve translation-equivariant image quantization by enforcing orthogonality among the codebook embeddings. To explore the advantages of translation-equivariant image quantization, we conduct three proof-of-concept experiments with a carefully controlled dataset: (1) text-to-image generation, where the quantized image indices are the target to predict, (2) image-to-text generation, where the quantized image indices are given as a condition, (3) using a smaller training set to analyze sample efficiency. From the strictly controlled experiments, we empirically verify that the translation-equivariant image quantizer improves not only sample efficiency but also the accuracy over VQGAN up to +11.9% in text-to-image generation and +3.9% in image-to-text generation.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 15 of its 22 references that are in the map. First posted on arXiv 1 Dec 2021.

Builds on

15 of 22 references.

References · 15

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  5. 2017Neural Discrete Representation Learningcited by 4,650
  6. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  7. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  8. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  9. 2021Alias-Free Generative Adversarial Networkscited by 1,181
  10. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 691
  11. 2021Vector-quantized Image Modeling with Improved VQGANcited by 656
  12. 2019Making Convolutional Networks Shift-Invariant Againcited by 390
3 more
  1. 2018Why do deep convolutional networks generalize so poorly to small image transformations?cited by 198
  2. 2020Delving Deeper into Anti-aliasing in ConvNetscited by 60
  3. 2021Impact of Aliasing on Generalization in Deep Convolutional Networkscited by 25

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 4; 2025: 2; 2026: 3.

10 papers.

  1. 2024GFT: Graph Foundation Model with Transferable Tree Vocabularycited by 82
  2. 2023ADDP: Learning General Representations for Image Recognition and Generation with Alternating Denoising Diffusion Processcited by 19
  3. 2024Representation Collapsing Problems in Vector Quantizationcited by 10
  4. 2024Identifying Spatio-Temporal Drivers of Extreme Eventscited by 1
  5. 2026VP-VAE: Rethinking Vector Quantization via Adaptive Vector Perturbationcited by 1
  6. 2026Beyond Stationarity: Rethinking Codebook Collapse in Vector Quantizationcited by 1
  7. 2026Early Quantization Shrinks Codebook: A Simple Fix for Diversity-Preserving Tokenizationcited by 1
  8. 2024Identifying and Clustering Counter Relationships of Team Compositions in PvP Games for Efficient Balance Analysiscited by 0
  9. 2025VQEL: Enabling Self-Play in Emergent Language Games via Agent-Internal Vector Quantizationcited by 0
  10. 2025Hierarchical Spatial Algorithms for High-Resolution Image Quantization and Feature Extractioncited by 0