Study finds single confidence thresholds yield unequal coverage across languages

A new arXiv paper examines whether one confidence threshold gives equal human-review coverage across languages in typed decision interfaces, finding gaps that shrink when inputs are translated.

Read the full story at arXiv cs.CL ↗

Timeline · 1 report

1. 2026-10-30 04:00 · arXiv cs.CL
   One Confidence Threshold, Unequal Coverage: Language Gaps in Typed Decision Interfaces and How Translating Inputs Reduces Them