TriviaQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TriviaQA 30K context
0.813Accuracy
18
TriviaQA
78.4F1 Score
17
TriviaQA
65.7EM
17
TriviaQA
70.31Exact Match (EM)
17
TriviaQA
74Accuracy (%)
17
TriviaQA
1.597d'type2
17
TriviaQA
78.4F1 Score
17
TriviaQA In-domain (test)
99.1Accuracy
16
TriviaQA
72.8Exact Match (EM)
16
TriviaQA fixed split (test)
66.41EM
16
TriviaQA
64.6Score
16
TriviaQA
0.766PR-AUC
16
TriviaQA
77Accuracy
16
TriviaQA
0.909Precision@1
16
TriviaQA v1.0 (test)
0.688A1 Accuracy
16
TriviaQA (TQA) (test)
79.8Exact Match
16
TriviaQA OOD
67.5Exact Match Accuracy
15
TriviaQA Gemini outputs (test)
0.67AUROC
15
TriviaQA GPT outputs (test)
0.675AUROC
15
TriviaQA Llama outputs (test)
0.844AUROC
15
TriviaQA Out-of-distribution (test)
72.64Accuracy
15
TriviaQA
89.48AUC
15
TriviaQA
0.131Mean Per-Step Regret
15
TriviaQA
73.4Accuracy (Benign)
15
TriviaQA
80.09Task Performance
14