TriviaQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TriviaQA Single-Answer
1.89ECE
11
TriviaQA
77Accuracy
11
TriviaQA (test)
66.1F1
11
TriviaQA
77.3TriviaQA Score
11
TriviaQA
88.26Accuracy
11
TriviaQA Web domain Verified (test)
82.4Exact Match (EM)
11
TriviaQA S→T
42.5Accuracy
10
TriviaQA (test)
60EM
10
TriviaQA
5.576Overall Score
10
TriviaQA
80LLM-as-a-Judge Score
10
TriviaQA
0.75Cohen's d
10
TriviaQA
64.6Exact Match (EM) Accuracy
10
TriviaQA
66.3EM
10
TriviaQA
83.9Exact Match (EM)
10
TriviaQA
72.17F1 Score
10
TriviaQA
79.8EM
10
TriviaQA
72.93Score
10
TriviaQA single-hop (test)
63.9Average Rate @16
10
TriviaQA
61.61EM
10
TriviaQA 5-shot
82.17Exact Match Accuracy
10
TriviaQA (TQA) (test)
64.5EM
10
TriviaQA
0.51Recall@1
10
TriviaQA (TQA) top 1000 samples (test)
52.3EM
10
TriviaQA (test)
72.67Exact Match (EM)
10
TriviaQA (TQ) (dev)
0.726Exact Match (EM)
10