Datasets:
Tasks:
Text Classification
Size:
100K<n<1M
Tags:
machine-generated
diacritization
african-languages
data-quality
synthetic-degradation
llm-as-judge
License:
Commit History
5 days ago
-
493,831 rows, 71 languages, synthetic degradation + LLM-judge prompts
Verified signatureab21aaf -
284,013 rows, 71 languages, synthetic degradation + LLM-judge prompts
Verified signatureba03564 -
71,775 rows, 69 languages, synthetic degradation + LLM-judge prompts
Verified signatureb8155a8