Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
97.95Attack Success Rate (ASR)
20
Feb 26, 2026
50.38TruthfulQA Score
20
Feb 26, 2026
75.8EM
20
Feb 26, 2026
42.2EM
20
Feb 26, 2026
52.41F1 Score
20
Feb 26, 2026
42.6Accuracy
20
Feb 26, 2026
69.5Accuracy
20
Jun 23, 2026
89.4F1 Score
20
Feb 26, 2026
80.9F1
20
Feb 26, 2026
34.7EM
20
Feb 26, 2026
82.4Accuracy
19
Jun 18, 2026
75.3Accuracy
19
Jul 9, 2026
42.5EM
19
May 19, 2026
84.21QA Score (TR Context)
19
May 18, 2026
57.6EM
19
May 12, 2026
82.67Accuracy
19
Apr 14, 2026
52.3FAR (Overall)
19
Apr 7, 2026
50.1ROUGE-L
19
May 26, 2026
54.57F1 Score
19
Apr 3, 2026
59.1F1 Score
19
Apr 3, 2026
80.8Accuracy
19
Feb 27, 2026
14.5EM
19
Feb 26, 2026
69.8Accuracy
19
Mar 18, 2026
96.2Accuracy
19
Jun 4, 2026
74.6Accuracy
19
Feb 27, 2026