Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
91.8Accuracy
3
Mar 24, 2026
25.969Perplexity
3
Mar 20, 2026
15.015Perplexity
3
Mar 20, 2026
60.4Accuracy
3
Mar 20, 2026
80.1Normalized Log Accuracy
3
Mar 18, 2026
31.5Precision
3
Mar 18, 2026
32.9Precision
3
Mar 18, 2026
2.92Average Score
3
Mar 17, 2026
17.75MultiFieldQA Score
3
Mar 12, 2026
24.5EM-R
3
Mar 6, 2026
39.7EM-R
3
Mar 6, 2026
66.7Trivial Score
3
Mar 5, 2026
0.115ECE
3
Mar 4, 2026
83.67Single Choice Accuracy
3
Feb 27, 2026
0.888Accuracy
3
Feb 26, 2026
75.9Accuracy
3
Feb 26, 2026
63.6Accuracy
3
Feb 26, 2026
80.6Accuracy
3
Feb 26, 2026
67.3Accuracy
3
Feb 26, 2026
0.012Loss
3
Feb 26, 2026
71.26Pattern-based Score
3
Feb 26, 2026
30BLEU-1
3
Jun 30, 2026
60.8EM
3
Feb 26, 2026
91.9ROUGE-L Recall
3
Feb 26, 2026
81.62Accuracy
3
Feb 26, 2026