# Paraphrase Identification on QQP, PAWS, and PIT (In-distribution test) Source: [Paraphrase Identification on QQP, PAWS, and PIT (In-distribution test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:26:26.511Z Evaluation of model performance on the in-distribution test sets of the Quora Question Pairs (QQP), PAWS, and PIT datasets for paraphrase identification, reporting AUROC scores for intra- and cross-dataset evaluations. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Paraphrase Identification | | Dataset (including any stated split) | QQP, PAWS, and PIT In-distribution (test) | | Metric | AUROC (QQP-\>QQP) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [AUROC (QQP-\>QQP)](): Higher is better; unit: PERCENT. - [AUROC (PAWS-\>PAWS)](): Higher is better; unit: PERCENT. - [AUROC (PIT-\>PIT)](): Higher is better; unit: PERCENT. - [AUROC (QQP-\>PIT)](): Higher is better; unit: PERCENT. - [AUROC (PIT-\>QQP)](): Higher is better; unit: PERCENT. - [Average AUROC](): Higher is better; unit: PERCENT. ## Recorded results | Method | AUROC (QQP-\>QQP) | Result date | Report paper | | --- | --- | --- | --- | | ROBERTa | 92 | Not recorded | 68fa6d931e209131cea3ac4f | | GAPX | 90.7 | Not recorded | 68fa6d931e209131cea3ac4f | | BART | 90.6 | Not recorded | 68fa6d931e209131cea3ac4f | | BERT | 90 | Not recorded | 68fa6d931e209131cea3ac4f | | BERT+EP | 89.8 | Not recorded | 68fa6d931e209131cea3ac4f | | IDP | 88.1 | Not recorded | 68fa6d931e209131cea3ac4f | | OODP | 79.6 | Not recorded | 68fa6d931e209131cea3ac4f | | GAP | 78.9 | Not recorded | 68fa6d931e209131cea3ac4f | | BILSTM | 64.6 | Not recorded | 68fa6d931e209131cea3ac4f | | BOW | 60.9 | Not recorded | 68fa6d931e209131cea3ac4f | ### ROBERTa AUROC (QQP-\>QQP): 92. Result ID: 698faba13a6c314849faf756. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GAPX AUROC (QQP-\>QQP): 90.7. Result ID: 698faba13a6c314849faf75a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BART AUROC (QQP-\>QQP): 90.6. Result ID: 698faba13a6c314849faf755. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BERT AUROC (QQP-\>QQP): 90. Result ID: 698faba13a6c314849faf753. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BERT+EP AUROC (QQP-\>QQP): 89.8. Result ID: 698faba13a6c314849faf754. [Reporting paper]() Recorded evaluation configuration: {"Entropy Penalty":"true"}. This record does not assert independent reproduction. ### IDP AUROC (QQP-\>QQP): 88.1. Result ID: 698faba13a6c314849faf758. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### OODP AUROC (QQP-\>QQP): 79.6. Result ID: 698faba13a6c314849faf757. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GAP AUROC (QQP-\>QQP): 78.9. Result ID: 698faba13a6c314849faf759. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BILSTM AUROC (QQP-\>QQP): 64.6. Result ID: 698faba13a6c314849faf752. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BOW AUROC (QQP-\>QQP): 60.9. Result ID: 698faba13a6c314849faf751. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.