List your company

ResearchFormerly WizWand

Paraphrase Identification

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

91.7Accuracy

78

Feb 26, 2026

Quora Question Pairs (test)

91.3Accuracy

72

Feb 26, 2026

91.7Accuracy

66

Apr 15, 2026

0.37G(W) Error Approximation

40

Jun 15, 2026

97.6Accuracy

35

Jun 26, 2026

CaliDist_verbalized (Sa.)

2.28ECE

32

Jun 5, 2026

-4.73Delta 1

24

Feb 26, 2026

TwitterPara (test)

SBERT-supervised

81.6TURL

22

Feb 26, 2026

PAWS -> PAWS (test)

94.1Accuracy

22

Feb 26, 2026

91.4Accuracy

22

Apr 28, 2026

QQP Out-of-distribution from PAWS

70.8Macro F1

20

Feb 26, 2026

PAWS -> QQP (test)

77.7AUROC

20

Feb 26, 2026

PIT -> PAWS (test)

77.5AUROC

20

Feb 26, 2026

PAWS-X 1.0 (test)

55.3Accuracy (de)

18

May 14, 2026

QQP few-shot zero-shot

74Accuracy

16

Feb 26, 2026

MCEval PAWS 8K (test)

88.9Accuracy

14

Feb 26, 2026

Quora Question Pairs

95.3Accuracy

14

Feb 26, 2026

Quora Question Pairs (dev)

90.48Accuracy

14

Feb 26, 2026

mBERT (Reference [1])

94Accuracy (en)

13

Feb 27, 2026

54.6Accuracy (de)

12

Feb 26, 2026

WMT Out-of-distribution from PAWS

76.4Macro F1

10

Feb 26, 2026

PIT Out-of-distribution from PAWS

69.8Macro F1

10

Feb 26, 2026

PAWS Out-of-distribution from PIT

55.1Macro F1

10

Feb 26, 2026

QQP Out-of-distribution from PIT

0.757Macro F1

10

Feb 26, 2026

WMT Out-of-distribution from QQP

75.5Macro F1

10

Feb 26, 2026