HotpotQA
Benchmarks
TaskDatasetSOTA resultTrendResults
HotpotQA
0.86Spearman Correlation
1
HotpotQA Large-scale (train)
2.1# Search Steps
1
HotpotQA leaderboard official (test)
73.62Overall Score
1
HotpotQA leaderboard official (test)
0.891Score
1
HotpotQA Extreme (test)
—F1 Score
0
HotpotQA Noisy (test)
—F1
0
HotpotQA Normal (test)
—F1
0
HotpotQA
—Score
0
HotpotQA in-distribution v4 (test)
—EM
0
HotpotQA D (val)
—Accuracy
0
HotpotQA D^v (train)
—Accuracy
0