Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1.594CPR
4
Jun 11, 2026
33.98Accuracy (SciQ)
4
Jun 8, 2026
85EM Score
4
Jun 2, 2026
71EM
4
Jun 2, 2026
66Exact Match (EM)
4
Jun 2, 2026
60.79Accuracy
4
Jun 2, 2026
32.7EM
4
Jun 2, 2026
97.8Accuracy
4
May 28, 2026
69Accuracy
4
May 28, 2026
55.6Accuracy
4
May 28, 2026
7.08Reasoning Score
4
May 26, 2026
42.5Accuracy (Before)
4
May 26, 2026
89.5Average Accuracy
4
May 26, 2026
88Accuracy
4
May 26, 2026
89.9Accuracy
4
May 26, 2026
90.7Accuracy
4
May 26, 2026
79.49Accuracy
4
May 26, 2026
30.17ROUGE-L
4
May 26, 2026
39.2Evidence Score
4
May 25, 2026
58.2Evidence Score
4
May 25, 2026
48.4Evidence Score
4
May 25, 2026
8.2Risk
4
May 21, 2026
65.1Accuracy
4
May 19, 2026
74.1Accuracy
4
May 19, 2026
0.697Cosine Similarity
4
May 19, 2026