Metraipolis › archive › dc_f38fdf0d92a2

Can Scene Text Recognition Read Rare Compositions? — arXiv:2609.00816

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.00816
License (as archived)
CC0-1.0
Archived
2026-10-17T07:13:38.594Z
SHA-256 of the archived text
57787165863b8805a44e571c17d7c2ef6da35d5db6cb48bd33203e79ab028f49
Stable URL of this page
https://metraipolis.com/documento/dc_f38fdf0d92a2

How to cite

Can Scene Text Recognition Read Rare Compositions? — arXiv:2609.00816 — arXiv (en), https://arxiv.org/abs/2609.00816 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f38fdf0d92a2 (Metraipolis archive dc_f38fdf0d92a2, sha256 57787165863b8805a44e571c17d7c2ef6da35d5db6cb48bd33203e79ab028f49)
BibTeX
@misc{dcf38fdf0d92a2,
  title = {Can Scene Text Recognition Read Rare Compositions? — arXiv:2609.00816},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.00816},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f38fdf0d92a2; sha256 57787165863b8805a44e571c17d7c2ef6da35d5db6cb48bd33203e79ab028f49},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00816. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Scene text recognition is reported as 89--97% accurate on the six standard benchmarks, and the problem is widely treated as saturated. We present an alternative reading. When the same test images are stratified jointly by ground-truth word rarity and character n-gram novelty against a reference corpus, accuracy at the rare-word x rare-trigram corner of the resulting 5x5 grid drops 10--18 pt below the q3/q3 centre across nine English specialised recognisers, and the same direction (corner below centre) holds on all 13 of 13 (language, model) pairs we test across four writing systems (Latin, Han, Han+kana, Arabic). The drop is not a capacity bottleneck. A 6x vision-backbone scale-up (CLIP4STR-Base 158M -> CLIP4STR-Huge 1.0B, OpenCLIP ViT-H/14 LAION-2B) leads every benchmark in aggregate accuracy yet leaves the stress corner unchanged (86.9 -> 86.5, within paired-bootstrap noise). Four converging probes--layer-wise probing, confidence-when-wrong, attention re-balancing, and a cross-script commit-vs-abstain error split--localise the failure to the autoregressive decoder's lexical prior. We then ask how much of the gap existing techniques recover. Of 16 non-architectural mitigations, the largest mean q5/q5 gain is +1.3 pt and none clears the paired-bootstrap noise floor; the only intervention that does is the architectural shift from autoregressive to CTC decoding (SVTRv2, +2.5 pt, p=0.02, n=474). A confidence-routed AR-CTC ensemble adds a directionally consistent +0.6 pt that stays within noise, and its dominant learned coefficient is each model's own minimum-softmax confidence--independently echoing the mechanism above. No configuration we test improves both the compositional corner and aggregate accuracy. The rare-input long tail thus points to architectural change rather than added capacity.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.00816 — 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 2026-10-17T07:00:58.936Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).