Datasets:
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
100K - 1M
ArXiv:
Tags:
multi-hop-question-answering
hotpotqa
evidence-selection
question-decomposition
chain-of-thought
supervised-fine-tuning
License:
Commit History
-
Refine dataset artwork and pipeline presentation
Verified signature499f1c2 -
Polish dataset card and add centered project artwork
Verified signatured5b707c -
Polish dataset card and add centered project artwork
Verified signature43d0573
-
Pin the immutable release revision in the dataset card
Verified signaturef5a70de -
Publish complete validated Bactrainus HotpotQA training data
Verified signature7f3a1d4 -
Add guarded final publisher and document two-stage validation
Verified signatureca40870 -
Publish validated release tooling and transparent pending-data card
Verified signature93a180e -
initial commit
Verified signaturee3ee179