Backdoor Attack on Fear Trigger Emotion (OOD)
81.3ASR (OOD)GREAT
Evaluation Results
| Method | Links | |
|---|---|---|
| GREATModel=Llama-3.2-1B, Poisoning ratio (α)=10%, K=20002025.10 | 81.3 | |
| GREATModel=Llama-3.2-3B, Poisoning ratio (α)=10%, K=20002025.10 | 77.2 | |
| GREATModel=Llama-3.2-3B, Poisoning ratio (α)=1%, K=1002025.10 | 76.3 | |
| GREATModel=Llama-3.2-3B, Poisoning ratio (α)=10%, K=12025.10 | 62.2 | |
| RandomModel=Llama-3.2-3B, Poisoning ratio (α)=10%, K=20002025.10 | 58.3 | |
| RandomModel=Llama-3.2-1B, Poisoning ratio (α)=10%, K=20002025.10 | 57.2 | |
| GREATModel=Llama-3.2-1B, Poisoning ratio (α)=10%, K=12025.10 | 57 | |
| RandomModel=Llama-3.2-3B, Poisoning ratio (α)=1%, K=1002025.10 | 56.7 | |
| GREATModel=Llama-3.2-1B, Poisoning ratio (α)=1%, K=1002025.10 | 56.2 | |
| GREATModel=Llama-3.2-1B, Poisoning ratio (α)=1%, K=12025.10 | 54 | |
| RandomModel=Llama-3.2-1B, Poisoning ratio (α)=10%, K=12025.10 | 53.5 | |
| RandomModel=Llama-3.2-3B, Poisoning ratio (α)=10%, K=12025.10 | 51.2 | |
| RandomModel=Llama-3.2-1B, Poisoning ratio (α)=1%, K=1002025.10 | 50.8 | |
| SUDOModel=Llama-3.2-3B, Poisoning ratio (α)=10%2025.10 | 49.1 | |
| GREATModel=Llama-3.2-3B, Poisoning ratio (α)=1%, K=12025.10 | 48.5 | |
| SUDOModel=Llama-3.2-1B, Poisoning ratio (α)=10%2025.10 | 48.2 | |
| RandomModel=Llama-3.2-3B, Poisoning ratio (α)=1%, K=12025.10 | 48 | |
| SUDOModel=Llama-3.2-1B, Poisoning ratio (α)=1%2025.10 | 47 | |
| RandomModel=Llama-3.2-1B, Poisoning ratio (α)=1%, K=12025.10 | 46.5 | |
| SUDOModel=Llama-3.2-3B, Poisoning ratio (α)=1%2025.10 | 45 |