Metraipolis › archive › dc_c71dd0c11fc7

One Confidence Threshold, Unequal Coverage: Language Gaps in Typed Decision Interfaces and How Translating Inputs Reduces Them — arXiv:2610.37688

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.37688
License (as archived)
CC0-1.0
Archived
2026-10-31T06:10:58.551Z
SHA-256 of the archived text
08ab88100a882dffdfea31982856ad06d1b8705a136cfa1f74acb33237190ded
Stable URL of this page
https://metraipolis.com/documento/dc_c71dd0c11fc7

How to cite

One Confidence Threshold, Unequal Coverage: Language Gaps in Typed Decision Interfaces and How Translating Inputs Reduces Them — arXiv:2610.37688 — arXiv (en), https://arxiv.org/abs/2610.37688 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c71dd0c11fc7 (Metraipolis archive dc_c71dd0c11fc7, sha256 08ab88100a882dffdfea31982856ad06d1b8705a136cfa1f74acb33237190ded)
BibTeX
@misc{dcc71dd0c11fc7,
  title = {One Confidence Threshold, Unequal Coverage: Language Gaps in Typed Decision Interfaces and How Translating Inputs Reduces Them — arXiv:2610.37688},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.37688},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c71dd0c11fc7; sha256 08ab88100a882dffdfea31982856ad06d1b8705a136cfa1f74acb33237190ded},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.37688. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deployed classifiers often act on a prediction only when its confidence clears a threshold and pass the rest to a person. We ask whether one threshold serves all languages equally on typed decision interfaces, APIs that return a probability for each option of a fixed multiple-choice question. In a confirmatory study with a written analysis plan, three vendors' interfaces classified 1,000 MASSIVE items, each in 51 languages, with a threshold of 0.9. Even when a request is answered correctly in both English and another language, the English version clears the threshold more often, by 8.6-11.7 percentage points on average. Observed coverage, the share of requests handled automatically, is lower than English's in every other language, while the requests the gate handles are rarely wrong. The gap remains when every language is reweighted to the same accuracy, and a standard calibration metric (ECE) does not detect it on one interface. In exploratory analyses, English's coverage stays higher when each language gets its own threshold fitted to the same 5% error target. Translating inputs into English nearly closes the gap at 0.9 but leaves English ahead at that error target. Requests written natively in African languages show a gap in the same direction when English is their human translation. We recommend reporting per-language coverage at the operating threshold.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.37688 — 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 2026-10-31T06:07:08.582Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).