🛰️ Daily AI Frontier
‹ back to 2026-09-23

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

Research Multimodal & Generative

Ranking

Overall 83
Content 90
Popularity 66

Observed public metrics from 1 member.

Representative image for StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

Merged summary

TL;DR - StableVQ is a parameter-free training approach for more stable vector-quantized visual tokenizers. It separates encoder-decoder and codebook responsibilities to improve codebook utilization and reconstruction quality across varied ImageNet settings.

  • Dynamic STE stabilizes encoder optimization under discrete regularization, including when codebook utilization is low.
  • Region VQ Loss enables the codebook to track the encoder output distribution without depending on encoder oscillations.
  • Independent learning-rate schedules reflect the different optimization dynamics of the encoder-decoder and codebook.
  • Built on shared-projection codebooks, StableVQ adds no learnable parameters.

Sources (1)

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

arXiv cs.CV Bao Tang, Jiahao Guo, Haoxiang Cao, Wenyu Liu, Changqian Yu, Kun Gai, Xinggang Wang 2026-09-22 arXiv:2609.26774
Public signals Hugging Face upvotes 30
Providers: Hugging Face · Upvotes 30 OpenAlex · N/A Publisher · N/A Semantic Scholar · N/A X · N/A Fetched 2026-09-25 14:16:48.020119 UTC

TL;DR - StableVQ is a parameter-free training approach for more stable vector-quantized visual tokenizers. It separates encoder-decoder and codebook responsibilities to improve codebook utilization and reconstruction quality across varied ImageNet settings.

  • Dynamic STE stabilizes encoder optimization under discrete regularization, including when codebook utilization is low.
  • Region VQ Loss enables the codebook to track the encoder output distribution without depending on encoder oscillations.
  • Independent learning-rate schedules reflect the different optimization dynamics of the encoder-decoder and codebook.
  • Built on shared-projection codebooks, StableVQ adds no learnable parameters.
item →