# Natural Language Understanding on GLUE (val) - Subset (SST-2, QNLI, QQP, MNLI) Source: [Natural Language Understanding on GLUE (val) - Subset (SST-2, QNLI, QQP, MNLI)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:03:31.296Z Evaluation of Natural Language Understanding capabilities focusing only on the SST-2, QNLI, QQP, and MNLI tasks from the GLUE validation set. 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 | Natural Language Understanding | | Dataset (including any stated split) | GLUE (val) | | Metric | SST-2 Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | 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 - [SST-2 Accuracy](): Higher is better; unit: PERCENT. - [QNLI Accuracy](): Higher is better; unit: PERCENT. - [QQP Accuracy](): Higher is better; unit: PERCENT. - [MNLI Accuracy](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | SST-2 Accuracy | Result date | Report paper | | --- | --- | --- | --- | | RoLoRA | 94.61 | Not recorded | 68fa79061e209131cea7bc39 | | FedTT+ | 94.5 | Not recorded | 68fa79061e209131cea7bc39 | | FedTT | 94.38 | Not recorded | 68fa79061e209131cea7bc39 | | LoRA | 93.23 | Not recorded | 68fa79061e209131cea7bc39 | | FFA-LoRA | 92.32 | Not recorded | 68fa79061e209131cea7bc39 | ### RoLoRA SST-2 Accuracy: 94.61. Result ID: 6995ba373badfaffe6bb869b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Rank":"2","Parameter Count":"34K","Data Distribution":"severe heterogeneity"}. This record does not assert independent reproduction. ### FedTT+ SST-2 Accuracy: 94.5. Result ID: 6995ba373badfaffe6bb869d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Parameter Count":"28K","Data Distribution":"severe heterogeneity"}. This record does not assert independent reproduction. ### FedTT SST-2 Accuracy: 94.38. Result ID: 6995ba373badfaffe6bb869c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Parameter Count":"51K","Data Distribution":"severe heterogeneity"}. This record does not assert independent reproduction. ### LoRA SST-2 Accuracy: 93.23. Result ID: 6995ba373badfaffe6bb8699. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Rank":"2","Parameter Count":"68K","Data Distribution":"severe heterogeneity"}. This record does not assert independent reproduction. ### FFA-LoRA SST-2 Accuracy: 92.32. Result ID: 6995ba373badfaffe6bb869a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Rank":"2","Parameter Count":"34K","Data Distribution":"severe heterogeneity"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.