This website uses cookies to ensure you get the best experience.
Learn more.
A company compares answer quality by language. Overall accuracy improves, but one low-volume language worsens sharply. Which reporting practice exposes the regression?
Published
- Remove the low-volume language from all evaluation sets.
- Report metrics by language alongside aggregate results and sample counts.
- Choose only examples the new model answers correctly.
- Combine all languages and report only total answer length.
Comments — 0
Sign in to comment