| arithmetic-reasoning-on-gsm8k | Llama-3.3-70B + CAPO | #67 | Accuracy: 73.73 |
| arithmetic-reasoning-on-gsm8k | Mistral-Small-24B + CAPO | #78 | Accuracy: 65.07 |
| arithmetic-reasoning-on-gsm8k | Qwen2.5-32B + CAPO | #83 | Accuracy: 60.2 |
| sentiment-analysis-on-sst-5-fine-grained | Llama-3.3-70B + CAPO | #1 | Accuracy: 62.27 |
| sentiment-analysis-on-sst-5-fine-grained | Mistral-Small-24B + CAPO | #25 | Accuracy: 60.2 |
| sentiment-analysis-on-sst-5-fine-grained | Qwen2.5-32B + CAPO | #26 | Accuracy: 59.07 |
| subjectivity-analysis-on-subj | Llama-3.3-70B + CAPO | #15 | Accuracy: 91.6 |
| subjectivity-analysis-on-subj | Qwen2.5-32B + CAPO | #16 | Accuracy: 91 |
| subjectivity-analysis-on-subj | Mistral-Small-24B + CAPO | #18 | Accuracy: 81.67 |
| text-classification-on-ag-news | Llama-3.3-70B + CAPO | #19 | Error: 11.2 |
| text-classification-on-ag-news | Qwen2.5-32B + CAPO | #20 | Error: 12.93 |
| text-classification-on-ag-news | Mistral-Small-24B + CAPO | #22 | Error: 15.7 |
| text-classification-on-bala-copa | Qwen2.5-32B + CAPO | #1 | Accuracy: 98.47 |
| text-classification-on-bala-copa | Llama-3.3-70B + CAPO | #2 | Accuracy: 98.27 |
| text-classification-on-bala-copa | Mistral-Small-24B + CAPO | #3 | Accuracy: 95.13 |