Targeted Poisoning (Transfer) on TruthfulQA Two options (Free-form)
52.89Benign F1 ScoreNSP
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| NSPModel=Qwen2.5-7B2025.05 | 52.89 | 52.67 | 48.15 | 47.27 | 50.52 | 49.97 | |
| AAPModel=Qwen2.5-7B2025.05 | 46.68 | 46.68 | 55.94 | 55.09 | 51.31 | 50.89 | |
| PARASITEModel=Deepseek-7B2025.05 | 45.31 | 45.05 | 56.25 | 56.25 | 55.28 | 50.65 | |
| NSPModel=Deepseek-7B2025.05 | 43.67 | 43.67 | 47.55 | 47.55 | 45.61 | 45.61 | |
| AAPModel=Deepseek-7B2025.05 | 42.35 | 41.94 | 51.36 | 50.73 | 46.86 | 46.34 | |
| PARASITEModel=Qwen2.5-7B2025.05 | 41.55 | 41.55 | 62.18 | 60.69 | 51.81 | 51.07 | |
| ManualModel=Qwen2.5-7B2025.05 | 37.33 | 37.33 | 58.55 | 58.36 | 47.94 | 47.85 | |
| ECLIPSEModel=Deepseek-7B2025.05 | 25.04 | 24.98 | 23.61 | 22.96 | 24.33 | 23.97 | |
| ECLIPSEModel=Qwen2.5-7B2025.05 | 24.73 | 25.44 | 25.3 | 25.17 | 25.02 | 25.31 | |
| ManualModel=Deepseek-7B2025.05 | 1 | 1 | 0.18 | 0.18 | 0.59 | 0.59 |