TriviaQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TriviaQA (TQ) (test)
83.3Top5 Accuracy
4
TriviaQA (TQ) (dev)
83.5Top-5 Accuracy
4
TriviaQA Entity Subset (TQAe) (dev)
64.8Accuracy
4
TriviaQA (test)
68.3Exact Match (EM)
4
TriviaQA Open (dev)
50.7Exact Match
4
TriviaQA (test)
71.4F1
4
TriviaQA Web domain Verified (dev)
60.31Exact Match (EM)
4
TriviaQA Web domain Distant Supervision (dev)
46.68EM
4
TriviaQA Wiki domain Verified (test)
50.51Exact Match (EM)
4
TriviaQA (test)
78.5Always-RAG Accuracy
3
TriviaQA
85.6Exact Match (EM)
3
TriviaQA (out-of-domain)
11.9F1 Score
3
TriviaQA
66.76EM
3
TriviaQA Search-o1 framework (50% held-out test)
69.6Accuracy
3
TriviaQA Speech
75.5Accuracy
3
TriviaQA
100Success Rate
3
TriviaQA
69.5Accuracy (Compression)
3
TriviaQA
56Accuracy
3
TriviaQA
66.79Accuracy
3
TriviaQA (test)
67.7Accuracy@5
3
TriviaQA Open-Domain wiki (dev)
75.8Accuracy
3
TriviaQA (dev)
66.9F1 Score
3
TriviaQA Web (test)
40.87EM
3
TriviaQA
40EM
2
TriviaQA
77LAMA Score
2