Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

MRPC

Benchmarks

Task NameDataset NameSOTA ResultTrend
Paraphrase DetectionMRPC
Avg Accuracy89.9
89
Sentence SimilarityMRPC (test)
F1 (micro)78.36
44
Paraphrase DetectionMRPC GLUE (val)
Accuracy85.54
27
Paraphrase DetectionMRPC
Delta Accuracy0
24
Paraphrase IdentificationMRPC
Delta 1-4.73
24
Paraphrase DetectionMRPC
Accuracy90.43
14
Paraphrase DetectionMRPC
Spearman Correlation (x100)30.87
12
Ranking correlation with full dataset evaluationMRPC
Kendall Correlation0.65
10
ClassificationMRPC (test)
Macro F181.2
9
ParaphraseNO-MRPC NLEBench (test)
Accuracy73.7
6
ClassificationMRPC
Accuracy86.52
6
Paraphrase DetectionMRPC (val)
F1 Score93.8
6
Paraphrase DetectionMRPC (dev)
F1 Score91
6
Semantic EquivalenceMRPC
Success Rate70
5
Paraphrase IdentificationMRPC
Accuracy76
5
Paraphrase DetectionMRPC GLUE (test)
F1 Score88
5
Natural Language InferenceMRPC
Accuracy0.736
5
Binary ClassificationMRPC
AUC77.77
5
Paraphrase DetectionMRPC
Δ Accuracy-0.04
3
Indirect Prompt Injection SanitizationMRPC
GCG ASR0.5
2
Indirect Prompt Injection AttackMRPC
ASR98.5
2
Natural Language UnderstandingMRPC (test)
Accuracy89.46
2
Text ClassificationMRPC GLUE
Accuracy93.32
2
Paraphrase IdentificationMRPC
F1 Score88.9
2
Indirect Prompt Injection DetectionMRPC
GCG Accuracy95
1
Showing 25 of 25 rows