Metraipolis › archive › dc_bc83bcfbe519

What Makes a Good Medical Image Tokenizer? Rethinking Reconstruction and Generation in Medical Image Tokenization — arXiv:2609.24691

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.24691
License (as archived)
CC0-1.0
Archived
2026-10-09T22:06:19.391Z
SHA-256 of the archived text
85b02ac55025704142a319abf5e867554ffc87f6c11be214684b4b263119ba59
Stable URL of this page
https://metraipolis.com/documento/dc_bc83bcfbe519

How to cite

What Makes a Good Medical Image Tokenizer? Rethinking Reconstruction and Generation in Medical Image Tokenization — arXiv:2609.24691 — arXiv (en), https://arxiv.org/abs/2609.24691 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bc83bcfbe519 (Metraipolis archive dc_bc83bcfbe519, sha256 85b02ac55025704142a319abf5e867554ffc87f6c11be214684b4b263119ba59)
BibTeX
@misc{dcbc83bcfbe519,
  title = {What Makes a Good Medical Image Tokenizer? Rethinking Reconstruction and Generation in Medical Image Tokenization — arXiv:2609.24691},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.24691},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bc83bcfbe519; sha256 85b02ac55025704142a319abf5e867554ffc87f6c11be214684b4b263119ba59},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.24691. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Latent diffusion models now dominate medical image generation, and every such pipeline rests on a \emph{tokenizer} that compresses images into the latent codes for image generation to operate on. Thereby, the tokenizer choice bounds every downstream task from reconstruction fidelity and generation quality to the representations available for downstream analysis. Yet, medical imaging pipelines routinely utilize tokenizers from natural imaging on the hypothesis that their behavior carries over. However, this is an assumption never tested in the medical imaging regime, where datasets are orders of magnitude smaller and images exhibit far lower inter-sample variance. We present a systematic evaluation of medical image tokenizers evaluating thirty configurations across ten model families on twelve datasets at three compression factors, spanning reconstruction, generation, latent geometry, downstream classification, and memorization. We find that (1) performance on image reconstruction and generation strongly correlate, unlike prior reports on natural images; (2) modern tokenizers use nearly all of their codebook entries, but still leave most of the latent space unused; (3) training-set memorization is mild and is further suppressed by stronger latent space compression; and (4) discrete quantization can largely preserve downstream classification, with lookup-free schemes being the main exception.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.24691 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-09T21:45:21.079Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).