Datasets:
Tasks:
Question Answering
Modalities:
Text
Formats:
parquet
Sub-tasks:
extractive-qa
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit History
Convert dataset to Parquet (#7)
0a3a8b7
Convert dataset sizes from base 2 to base 10 in the dataset card (#3)
5fe18c4
add dataset_info in dataset metadata
33c0018
remove dummmy data
3431e84
mariosasko
commited on
Align more metadata with other repo types (models,spaces) (#4607)
d04f25d
Refactor column mappings for question answering datasets (#4391)
18217d5
Remove config names as yaml keys (#4367)
e9cb4f1
Eval metadata batch 3: Reddit, Rotten Tomatoes, SemEval 2010, Sentiment 140, SMS Spam, Snips, SQuAD, SQuAD v2, Timit ASR (#4337)
5ac3c74
Sasha Luccioni
commited on