TriviaQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TriviaQA out-of-domain (val test)
70.9EM
33
TriviaQA
0.87AUROC
32
TriviaQA (dev)
79.5F1 Score (Full)
32
TriviaQA Gen (test)
76.2Accuracy
31
TriviaQA LongBench
91.7F1 Score
30
TriviaQA
76.4EM
30
TriviaQA 5shots
80Accuracy
30
TriviaQA
83.4AUROC
29
TriviaQA (test)
73Accuracy
29
TriviaQA (test)
28.9Unreliability Rate
28
TriviaQA (40% test)
91AUROC
28
TriviaQA
62.11Exact Match (EM)
28
TriviaQA
0.323E-AURC
28
TriviaQA
0.044ECE
28
TriviaQA
81.3Exact Match AUC
28
TriviaQA (TQA)
0.418Accuracy
28
TriviaQA
65.3EM (%)
27
TriviaQA 1,000 samples (val)
71.9AUROC
27
TriviaQA (test)
80.6F1 Score
27
TriviaQA (test)
73.3EM
27
TriviaQA
65.2Accuracy
26
TriviaQA
71.8Accuracy
26
TriviaQA (TQA) (test)
77.1Accuracy
26
EVOUNA-TriviaQA
100F1 Score
25
TriviaQA
62Accuracy
24