Paraphrase Detection
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.1Empirical Error G(W)
90
Jun 15, 2026
90.43Accuracy
90
Jun 15, 2026
89.9Avg Accuracy
89
Feb 26, 2026
76.2Accuracy
68
Jun 4, 2026
95.3Accuracy
51
Feb 26, 2026
80.4Accuracy
34
Feb 26, 2026
85.54Accuracy
27
Feb 26, 2026
0Delta Accuracy
24
Feb 26, 2026
94.6Accuracy
24
Feb 27, 2026
91.79Accuracy
23
Feb 26, 2026
89.5Accuracy
21
Feb 26, 2026
96Accuracy
16
Feb 26, 2026
79.2Accuracy
15
May 8, 2026
77.4Accuracy
15
Feb 27, 2026
71.5Accuracy
12
Feb 26, 2026
86.8Accuracy
12
Mar 24, 2026
30.87Spearman Correlation (x100)
12
Feb 26, 2026
1.98Generalization Error G(W)
10
Jun 15, 2026
80.2Accuracy
10
Jun 4, 2026
66.1Accuracy
10
Apr 3, 2026
89F1 Score
10
Feb 26, 2026
60.8Accuracy
9
Feb 27, 2026
91.89Exact Match (EM)
8
Jun 4, 2026
84.8Accuracy
8
Feb 26, 2026
71.2Average Accuracy
8
Feb 26, 2026