# Language Understanding on GLUE and SuperGLUE Source: [Language Understanding on GLUE and SuperGLUE]() Collection index: [llms.txt]() Data updated: 2026-06-11T07:55:24.592Z Evaluation of model performance across selected tasks from the combined GLUE and SuperGLUE benchmarks, reporting scores for SST2, WSC, WIC, and BoolQ. 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 | Language Understanding | | Dataset (including any stated split) | GLUE and SuperGLUE fine-tuning | | Metric | SST-2 Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 36 of 36 | | 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. - [WSC Score](): Higher is better; unit: PERCENT. - [WIC Score](): Higher is better; unit: PERCENT. - [BoolQ Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | SST-2 Accuracy | Result date | Report paper | | --- | --- | --- | --- | | FlatPO | 96.1 | Not recorded | 6a28f32790966ade919c168e | | SAM | 95.8 | Not recorded | 6a28f32790966ade919c168e | | GAM | 95.8 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 95.8 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 95.5 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 95.4 | Not recorded | 6a28f32790966ade919c168e | | SAM | 95.1 | Not recorded | 6a28f32790966ade919c168e | | GAM | 95.1 | Not recorded | 6a28f32790966ade919c168e | | GAM | 95 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 94.8 | Not recorded | 6a28f32790966ade919c168e | | SAM | 94.8 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 94.7 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 94.6 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 94.5 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 94.4 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 94.3 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 94 | Not recorded | 6a28f32790966ade919c168e | | FlatPO | 93.9 | Not recorded | 6a28f32790966ade919c168e | | GAM | 93.8 | Not recorded | 6a28f32790966ade919c168e | | GAM | 93.8 | Not recorded | 6a28f32790966ade919c168e | | SAM | 93.6 | Not recorded | 6a28f32790966ade919c168e | | GAM | 93.6 | Not recorded | 6a28f32790966ade919c168e | | SAM | 93.6 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 93.4 | Not recorded | 6a28f32790966ade919c168e | | SAM | 93.4 | Not recorded | 6a28f32790966ade919c168e | | SAM | 93.2 | Not recorded | 6a28f32790966ade919c168e | | GAM | 93.2 | Not recorded | 6a28f32790966ade919c168e | | GAM | 93.2 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 93.2 | Not recorded | 6a28f32790966ade919c168e | | GAM | 93.1 | Not recorded | 6a28f32790966ade919c168e | | SAM | 93 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 92.8 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 92.7 | Not recorded | 6a28f32790966ade919c168e | | SAM | 92.7 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 92.6 | Not recorded | 6a28f32790966ade919c168e | | Adam (Base) | 92.6 | Not recorded | 6a28f32790966ade919c168e | ### FlatPO SST-2 Accuracy: 96.1. Result ID: 6a2a68a0fbc373592c274139. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 95.8. Result ID: 6a2a68a0fbc373592c274137. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 95.8. Result ID: 6a2a68a0fbc373592c274138. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 95.8. Result ID: 6a2a68a0fbc373592c274141. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 95.5. Result ID: 6a2a68a0fbc373592c274136. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 95.4. Result ID: 6a2a68a0fbc373592c27413d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 95.1. Result ID: 6a2a68a0fbc373592c27413f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 95.1. Result ID: 6a2a68a0fbc373592c274140. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 95. Result ID: 6a2a68a0fbc373592c27413c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 94.8. Result ID: 6a2a68a0fbc373592c274131. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 94.8. Result ID: 6a2a68a0fbc373592c27413b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 94.7. Result ID: 6a2a68a0fbc373592c27412d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 94.6. Result ID: 6a2a68a0fbc373592c274135. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 94.5. Result ID: 6a2a68a0fbc373592c27413a. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 94.4. Result ID: 6a2a68a0fbc373592c274121. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 94.3. Result ID: 6a2a68a0fbc373592c27413e. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA-13B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 94. Result ID: 6a2a68a0fbc373592c274125. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### FlatPO SST-2 Accuracy: 93.9. Result ID: 6a2a68a0fbc373592c274129. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 93.8. Result ID: 6a2a68a0fbc373592c27412c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 93.8. Result ID: 6a2a68a0fbc373592c274134. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 93.6. Result ID: 6a2a68a0fbc373592c27412b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 93.6. Result ID: 6a2a68a0fbc373592c274130. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 93.6. Result ID: 6a2a68a0fbc373592c274133. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 93.4. Result ID: 6a2a68a0fbc373592c27412a. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 93.4. Result ID: 6a2a68a0fbc373592c27412f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 93.2. Result ID: 6a2a68a0fbc373592c27411f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 93.2. Result ID: 6a2a68a0fbc373592c274120. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 93.2. Result ID: 6a2a68a0fbc373592c274124. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 93.2. Result ID: 6a2a68a0fbc373592c274132. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### GAM SST-2 Accuracy: 93.1. Result ID: 6a2a68a0fbc373592c274128. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 93. Result ID: 6a2a68a0fbc373592c274127. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 92.8. Result ID: 6a2a68a0fbc373592c27412e. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA-7B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 92.7. Result ID: 6a2a68a0fbc373592c274122. [Reporting paper]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### SAM SST-2 Accuracy: 92.7. Result ID: 6a2a68a0fbc373592c274123. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prefix Tuning"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 92.6. Result ID: 6a2a68a0fbc373592c27411e. [Reporting paper]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"LoRA"}. This record does not assert independent reproduction. ### Adam (Base) SST-2 Accuracy: 92.6. Result ID: 6a2a68a0fbc373592c274126. [Reporting paper]() Recorded evaluation configuration: {"Model":"OPT-1.3B","PEFT":"Prompt Tuning"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.