Hugging Face's logo

Datasets:
google
/
xtreme

Tasks:
Multiple Choice
Question Answering
Token Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
multiple-choice-qa
extractive-qa
open-domain-qa
Languages:
Afrikaans
Arabic
Bulgarian
Size:
1M - 10M
ArXiv:
Tags:
parallel-sentence-retrieval
paraphrase-identification
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files
xet
 Community
11
xtreme
837 kB
  • 6 contributors
 History: 22 commits
Michael Ustaszewski
Update BUCC18 parsing: ignore quotation marks in CSV Reader to avoid empty source or target sentences
77b8cb9