TriviaQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TriviaQA
75.2Exact Match (EM)
5
TriviaQA temporal
2.3ECE
5
TriviaQA 50/50 calibration (test)
74.5CR (%)
5
TriviaQA
28Accuracy
5
TriviaQA sampled
1.081Average Latency (s)
5
TriviaQA
62.2EM
5
TriviaQA
0.853E-SC
5
TriviaQA
0.197E-SC Score
5
TriviaQA
64Accg
5
TriviaQA
76subEM
5
TriviaQA
50.1NDCG@10
5
TriviaQA Open-book
91.4Recall
5
TriviaQA RC
83.08Accuracy
5
TriviaQA 11.3K (test)
40.2EM
5
TriviaQA (test)
50Exact Match
5
TriviaQA
83.9EM
5
TriviaQA
81.43EM
5
TriviaQA
52.51Trivia QA Score
5
TriviaQA (val)
16.6Exact Match
5
TriviaQA MRQA few-shot
80.7F1 Score
5
TriviaQA (test)
72Accuracy
5
TriviaQA (test)
35.45QAE EM
5
TriviaQA Wikipedia Full (dev)
0.511EM
5
TriviaQA Web (All) (test)
66.37Exact Match
5
TriviaQA Web domain Distant Supervision (test)
67.5EM
5