TriviaQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TriviaQA
0.714PRR (AlignScore)
14
TriviaQA
76.8T Score
14
TriviaQA (val)
92AUARC (Token Entropy)
14
TriviaQA (val)
0.83AUROC (Token Entropy)
14
TriviaQA 500 QA pairs
55.4Exact Match (EM)
14
TriviaQA 2018 Wikipedia dump (dev)
66.6Accuracy
14
TriviaQA Wikipedia dump December 2018 (test)
67.3EM
14
TriviaQA
78.75Recall
13
TriviaQA
12.2OSR (%)
13
TriviaQA
65.8EM
13
TriviaQA
56.7Accuracy
13
TriviaQA
71.8EM
12
TriviaQA held-out 50% (test)
0.939AUROC
12
TriviaQA
0.75Cohen's d
12
TriviaQA
0.518Spearman's rho
12
TriviaQA
76.3CEM
12
TriviaQA 200 samples (test)
62.5Rejection Accuracy (80%)
12
TriviaQA
85Hit@20
12
TriviaQA (test)
0.923F1 Score
12
TriviaQA in-corpus setting (test)
72ACC
12
TriviaQA
72.35Performance
12
TriviaQA
82Accuracy
12
TriviaQA
45.1Accuracy
11
TriviaQA matched-N (test)
88.1Accuracy
11
TriviaQA
66.6EM
11