Adversarial Attack Defense on GCG-Transfer
99.3BARVicuna-7B-chat-HF
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Vicuna-7B-chat-HFAlignment=Original LLM2023.09 | 99.3 | 83.3 | — | |
| Vicuna-7B-chat-HFAlignment=RA-LLM2023.09 | 98.7 | 11.3 | 71 | |
| Guanaco-7B-HFAlignment=Original LLM2023.09 | 95.3 | 78.7 | — | |
| Guanaco-7B-HFAlignment=RA-LLM2023.09 | 92 | 8.7 | 70 |