Robustness on Robustness
87.97RRSAC Multi-task
Evaluation Results
| Method | Links | |
|---|---|---|
| SAC Multi-taskModel=Qwen2-72B (Instruct)2024.11 | 87.97 | |
| SAC Single-taskModel=Qwen2-7B (Instruct)2024.11 | 85.89 | |
| SAC Single-taskModel=Qwen2-72B (Instruct)2024.11 | 84.23 | |
| SAC Multi-taskModel=Qwen2-7B (Instruct)2024.11 | 82.16 | |
| SAC Single-taskModel=Llama2-13B (Chat)2024.11 | 78.42 | |
| SAC Multi-taskModel=Llama2-13B (Chat)2024.11 | 75.93 | |
| No ControlModel=Qwen2-72B (Instruct)2024.11 | 58.51 | |
| No ControlModel=Qwen2-7B (Instruct)2024.11 | 57.68 | |
| No ControlModel=Llama2-13B (Chat)2024.11 | 39.42 |