Text Classification on CR
93.48CAAPO
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| APODataset-free=false2026.02 | 93.48 | — | — | — | |
| APOBackbone=Qwen2.5-7B-Instruct2025.05 | 93.48 | — | — | — | |
| DEDataset-free=false2026.02 | 93.38 | — | — | — | |
| DEBackbone=Qwen2.5-7B-Instruct2025.05 | 93.38 | — | — | — | |
| APEDataset-free=false2026.02 | 92.87 | — | — | — | |
| APEBackbone=Qwen2.5-7B-Instruct2025.05 | 92.87 | — | — | — | |
| PRLDataset-free=false2026.02 | 92.83 | — | — | — | |
| PRLBackbone=Qwen2.5-7B-Instruct, Prompt Selection=True2025.05 | 92.83 | — | — | — | |
| GADataset-free=false2026.02 | 92.75 | — | — | — | |
| TATRADataset-free=true2026.02 | 92.75 | — | — | — | |
| GABackbone=Qwen2.5-7B-Instruct2025.05 | 92.75 | — | — | — | |
| PIAST (E)Dataset-free=false2026.02 | 92.55 | — | — | — | |
| PIASTDataset-free=false2026.02 | 92.35 | — | — | — | |
| PRLBackbone=Qwen2.5-7B-Instruct, Prompt Selection=False2025.05 | 92.17 | — | — | — | |
| NIDataset-free=true2026.02 | 91.5 | — | — | — | |
| NIBackbone=Qwen2.5-7B-Instruct2025.05 | 91.5 | — | — | — | |
| STRIPAttack=AddSent, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 95.7 | 10.1 | 62.2 | |
| ONIONAttack=AddSent, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 95.7 | 4.75 | 19.5 | |
| RAPAttack=AddSent, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 95.7 | 12.95 | 48.9 | |
| MDPAttack=AddSent, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 95.7 | 4.8 | 3 | |
| STRIPAttack=SOS, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 100 | 2.2 | 78.2 | |
| ONIONAttack=SOS, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 100 | 0.75 | 37.55 | |
| RAPAttack=SOS, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 100 | 3.4 | 55.3 | |
| MDPAttack=SOS, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 91.45 | 100 | 0.2 | 0 | |
| GRACEDataset-free=false2026.02 | 90.92 | — | — | — | |
| GRACEBackbone=Qwen2.5-7B-Instruct2025.05 | 90.92 | — | — | — | |
| AncSetFitk=8, parameters=110M2026.01 | 90.7 | — | — | — | |
| GPS-JDataset-free=true2026.02 | 90.65 | — | — | — | |
| GPS-SR-0.1Dataset-free=true2026.02 | 90.5 | — | — | — | |
| GliClass 3.0k=0, parameters=151M2026.01 | 90.2 | — | — | — | |
| Curate-Train-Refinek=4, parameters=110M2026.01 | 90.1 | — | — | — | |
| STRIPAttack=BadNets, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.95 | 92.3 | 2.85 | 98.7 | |
| ONIONAttack=BadNets, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.95 | 92.3 | 5.2 | 7.45 | |
| RAPAttack=BadNets, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.95 | 92.3 | 1.35 | 43.6 | |
| MDPAttack=BadNets, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.95 | 92.3 | 4.95 | 5.1 | |
| AncSetFitk=4, parameters=110M2026.01 | 89.8 | — | — | — | |
| STRIPAttack=LWP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.75 | 91.3 | 1.8 | 99.1 | |
| ONIONAttack=LWP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.75 | 91.3 | 4.9 | 27.85 | |
| RAPAttack=LWP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.75 | 91.3 | 4.05 | 39.2 | |
| MDPAttack=LWP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.75 | 91.3 | 5.1 | 3.5 | |
| PromptAgentBackbone=Qwen2.5-7B-Instruct2025.05 | 89.68 | — | — | — | |
| PromptWizardBackbone=Qwen2.5-7B-Instruct2025.05 | 89.45 | — | — | — | |
| Curate-Train-Refinek=8, parameters=110M2026.01 | 89.4 | — | — | — | |
| STRIPAttack=EP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.35 | 67.55 | 2.2 | 87.2 | |
| ONIONAttack=EP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.35 | 67.55 | 10.15 | 4.4 | |
| RAPAttack=EP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.35 | 67.55 | 7.65 | 45.2 | |
| MDPAttack=EP, Learning Strategy=prompt-based fine-tuning, FRR threshold=5%2023.09 | 89.35 | 67.55 | 5.35 | 9.4 | |
| Curate-Train-Refinek=2, parameters=110M2026.01 | 89.01 | — | — | — | |
| Curate-Train-Refinek=0, parameters=110M2026.01 | 87.8 | — | — | — | |
| MIDataset-free=true2026.02 | 87.25 | — | — | — | |
| MIBackbone=Qwen2.5-7B-Instruct2025.05 | 87.25 | — | — | — | |
| AncSetFitk=2, parameters=110M2026.01 | 86.6 | — | — | — | |
| ADAPET-Basek=8, parameters=117M2026.01 | 78.6 | — | — | — | |
| ADAPET-Basek=4, parameters=117M2026.01 | 77 | — | — | — | |
| ADAPET-Basek=2, parameters=117M2026.01 | 76.8 | — | — | — |