# Text Classification on GLUE (SST2, QNLI, MNLI subset) Source: [Text Classification on GLUE (SST2, QNLI, MNLI subset)]() Collection index: [llms.txt]() Data updated: 2026-05-12T15:12:47.616Z Evaluation of Text Classification performance specifically on the SST-2, QNLI, and MNLI tasks within the GLUE benchmark suite. 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 | Text Classification | | Dataset (including any stated split) | GLUE | | Metric | Accuracy (SST-2) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 33 of 33 | | 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 - [Accuracy (SST-2)](): Higher is better; unit: PERCENT. - [Accuracy (QNLI)](): Higher is better; unit: PERCENT. - [Accuracy (MNLI)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (SST-2) | Result date | Report paper | | --- | --- | --- | --- | | DP-LAC (w/out Chist) | 89.1 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-CLAC | 86.4 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-LAC | 86.4 | Not recorded | 6a02b49edbc836b87d824e5d | | Bu (2023) | 84.8 | Not recorded | 6a02b49edbc836b87d824e5d | | Abadi (2016) | 84.6 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-LAC (w/out Chist) | 84.4 | Not recorded | 6a02b49edbc836b87d824e5d | | Bu (2023) | 83.5 | Not recorded | 6a02b49edbc836b87d824e5d | | Du (2022) | 83.4 | Not recorded | 6a02b49edbc836b87d824e5d | | Abadi (2016) | 83 | Not recorded | 6a02b49edbc836b87d824e5d | | Andrew (2021) | 82.7 | Not recorded | 6a02b49edbc836b87d824e5d | | Qiu (2024) | 81.9 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-LAC | 78.1 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-CLAC | 73.7 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-LAC (w/out Chist) | 70.5 | Not recorded | 6a02b49edbc836b87d824e5d | | Du (2022) | 68.8 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-LAC | 67.7 | Not recorded | 6a02b49edbc836b87d824e5d | | Bu (2023) | 66.9 | Not recorded | 6a02b49edbc836b87d824e5d | | Abadi (2016) | 66.8 | Not recorded | 6a02b49edbc836b87d824e5d | | Abadi (2016) | 63.9 | Not recorded | 6a02b49edbc836b87d824e5d | | Bu (2023) | 63.6 | Not recorded | 6a02b49edbc836b87d824e5d | | Andrew (2021) | 62.1 | Not recorded | 6a02b49edbc836b87d824e5d | | Andrew (2021) | 61.5 | Not recorded | 6a02b49edbc836b87d824e5d | | Abadi (2016) | 57.9 | Not recorded | 6a02b49edbc836b87d824e5d | | Bu (2023) | 57.7 | Not recorded | 6a02b49edbc836b87d824e5d | | Du (2022) | 56.3 | Not recorded | 6a02b49edbc836b87d824e5d | | Abadi (2016) | 55.8 | Not recorded | 6a02b49edbc836b87d824e5d | | Bu (2023) | 55.8 | Not recorded | 6a02b49edbc836b87d824e5d | | Andrew (2021) | 55.7 | Not recorded | 6a02b49edbc836b87d824e5d | | Andrew (2021) | 55.3 | Not recorded | 6a02b49edbc836b87d824e5d | | DP-CLAC | 55 | Not recorded | 6a02b49edbc836b87d824e5d | | Andrew (2021) | 54.9 | Not recorded | 6a02b49edbc836b87d824e5d | | Qiu (2024) | 52.7 | Not recorded | 6a02b49edbc836b87d824e5d | | Qiu (2024) | 52.6 | Not recorded | 6a02b49edbc836b87d824e5d | ### DP-LAC (w/out Chist) Accuracy (SST-2): 89.1. Result ID: 6a03435247f34ce924e852fe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### DP-CLAC Accuracy (SST-2): 86.4. Result ID: 6a03435247f34ce924e85310. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### DP-LAC Accuracy (SST-2): 86.4. Result ID: 6a03435247f34ce924e85313. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### Bu (2023) Accuracy (SST-2): 84.8. Result ID: 6a03435247f34ce924e852fb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### Abadi (2016) Accuracy (SST-2): 84.6. Result ID: 6a03435247f34ce924e852f5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### DP-LAC (w/out Chist) Accuracy (SST-2): 84.4. Result ID: 6a03435247f34ce924e852fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### Bu (2023) Accuracy (SST-2): 83.5. Result ID: 6a03435247f34ce924e8530a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### Du (2022) Accuracy (SST-2): 83.4. Result ID: 6a03435247f34ce924e85307. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### Abadi (2016) Accuracy (SST-2): 83. Result ID: 6a03435247f34ce924e85301. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### Andrew (2021) Accuracy (SST-2): 82.7. Result ID: 6a03435247f34ce924e852f8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### Qiu (2024) Accuracy (SST-2): 81.9. Result ID: 6a03435247f34ce924e8530d. [Reporting paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### DP-LAC Accuracy (SST-2): 78.1. Result ID: 6a03435247f34ce924e85312. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### DP-CLAC Accuracy (SST-2): 73.7. Result ID: 6a03435247f34ce924e8530f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### DP-LAC (w/out Chist) Accuracy (SST-2): 70.5. Result ID: 6a03435247f34ce924e852fc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Du (2022) Accuracy (SST-2): 68.8. Result ID: 6a03435247f34ce924e85306. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### DP-LAC Accuracy (SST-2): 67.7. Result ID: 6a03435247f34ce924e85311. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Bu (2023) Accuracy (SST-2): 66.9. Result ID: 6a03435247f34ce924e85309. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### Abadi (2016) Accuracy (SST-2): 66.8. Result ID: 6a03435247f34ce924e85300. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### Abadi (2016) Accuracy (SST-2): 63.9. Result ID: 6a03435247f34ce924e852f4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### Bu (2023) Accuracy (SST-2): 63.6. Result ID: 6a03435247f34ce924e852fa. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### Andrew (2021) Accuracy (SST-2): 62.1. Result ID: 6a03435247f34ce924e852f7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### Andrew (2021) Accuracy (SST-2): 61.5. Result ID: 6a03435247f34ce924e85304. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"8"}. This record does not assert independent reproduction. ### Abadi (2016) Accuracy (SST-2): 57.9. Result ID: 6a03435247f34ce924e852ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Bu (2023) Accuracy (SST-2): 57.7. Result ID: 6a03435247f34ce924e85308. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Du (2022) Accuracy (SST-2): 56.3. Result ID: 6a03435247f34ce924e85305. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Abadi (2016) Accuracy (SST-2): 55.8. Result ID: 6a03435247f34ce924e852f3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Bu (2023) Accuracy (SST-2): 55.8. Result ID: 6a03435247f34ce924e852f9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Andrew (2021) Accuracy (SST-2): 55.7. Result ID: 6a03435247f34ce924e852f6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"Def. HP","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Andrew (2021) Accuracy (SST-2): 55.3. Result ID: 6a03435247f34ce924e85303. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### DP-CLAC Accuracy (SST-2): 55. Result ID: 6a03435247f34ce924e8530e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Andrew (2021) Accuracy (SST-2): 54.9. Result ID: 6a03435247f34ce924e85302. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. ### Qiu (2024) Accuracy (SST-2): 52.7. Result ID: 6a03435247f34ce924e8530c. [Reporting paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"4"}. This record does not assert independent reproduction. ### Qiu (2024) Accuracy (SST-2): 52.6. Result ID: 6a03435247f34ce924e8530b. [Reporting paper]() Recorded evaluation configuration: {"Hyperparameter Setting":"DP HPO","Privacy Budget (epsilon)":"2"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.