Scientific Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
9,032Avg Response Length
13
May 12, 2026
60Accuracy
12
Jun 2, 2026
62.4Accuracy
12
May 29, 2026
82.4Accuracy
12
May 29, 2026
3.19SampleQA Score
11
Feb 26, 2026
89.84Accuracy
9
May 15, 2026
21.42Exact Match Accuracy
9
Mar 5, 2026
69.96Accuracy
8
Feb 26, 2026
64.9Chemistry Accuracy
6
Feb 26, 2026
80ROUGE-L Function Score
5
Jul 9, 2026
1,163Latency (NoKV, ms)
5
Jun 11, 2026
38.25Pass@1 Accuracy
5
May 1, 2026
33.9IK-IK Score
5
Apr 30, 2026
54.7Accuracy
4
Apr 15, 2026
27Score
4
Mar 6, 2026
39.72Answer F1
3
Apr 28, 2026
4SC Baseline AUC
2
May 8, 2026
92.42pass@50
2
Feb 26, 2026
79.27Avg@5 Accuracy (Seeds)
1
Feb 26, 2026