TriviaQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TriviaQA Web (Full)
68.65EM
7
TriviaQA Wikipedia Verified (test)
72.8EM
7
TriviaQA
68.2Answer in Context
7
TriviaQA Non-Factoid
31.9GSM8K Score
6
TriviaQA Factoid
90.8ID Score
6
TriviaQA
55.4Accuracy (EN)
6
TriviaQA
85.7Prediction Score
6
TriviaQA
57.21Accuracy
6
TriviaQA (test)
81.23Exact Match (EM)
6
TriviaQA
45.1Text Score
6
TriviaQA
76.66Benign F1
6
TriviaQA
42.85Score
6
TriviaQA (Memorization)
52F1 Score (TriviaQA Mem)
6
TriviaQA
74.5Reverse QA Score
6
TriviaQA TIP
62.82Precision
6
TriviaQA abundant
73.4F1 Score
6
TriviaQA (medium)
71EM & F1 Score
6
TriviaQA (scarce)
68.3F1 Score
6
TriviaQA (TQA) few-shot Filtered (test)
77.3Exact Match (EM)
6
TriviaQA (val)
74.8FT F1
6
TriviaQA Wikipedia domain
81EM
6
TriviaQA wiki split (test)
71.2Accuracy
6
TriviaQA Wiki domain Distant Supervision (test)
64.6Exact Match
6
TriviaQA
89.2LAMA Score
5
TriviaQA out-of-domain (test)
60.7Accuracy
5