Negative Sentiment Backdoor Detection on Gemma 2 9B
0Attack Success Rate (ASR)
LCF
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.04 | 0 | - | 15 | |
2026.04 | 0 | - | 10.5 | |
2026.04 | 0 | - | 13.5 | |
2026.04 | 0 | - | 13 | |
2026.04 | 0 | - | 19 | |
2026.04 | 0 | - | 14 | |
2026.04 | 0 | - | 20.5 | |
2026.04 | 0.7 | - | 14.8 | |
2026.04 | 6.5 | 2.5 | - | |
2026.04 | 12 | 1.5 | - | |
2026.04 | 12.3 | 1.6 | - | |
2026.04 | 14.5 | 1.5 | - | |
2026.04 | 16.5 | 2 | - | |
2026.04 | 26 | 2 | - | |
2026.04 | 26.7 | 1 | - | |
2026.04 | 32 | 1.5 | - | |
2026.04 | 82 | 2.5 | - | |
2026.04 | 90 | 1.5 | - | |
2026.04 | 91.5 | 1 | - | |
2026.04 | 92.5 | 1.5 | - | |
2026.04 | 92.7 | 1.9 | - | |
2026.04 | 93.5 | 2 | - | |
2026.04 | 93.6 | 10 | - | |
2026.04 | 97.5 | 1 | - | |
2026.04 | 98.5 | 1.5 | - | |
2026.04 | 99 | 2 | - | |
2026.04 | 99.5 | 1.5 | - | |
2026.04 | 99.6 | 3.8 | - | |
2026.04 | 99.8 | 3 | - | |
2026.04 | 99.9 | 3.4 | - | |
2026.04 | 100 | 1.5 | - | |
2026.04 | 100 | 1.5 | - | |
2026.04 | 100 | 2 | - | |
2026.04 | 100 | 1.5 | - | |
2026.04 | 100 | 1.5 | - | |
2026.04 | 100 | 1 | - | |
2026.04 | 100 | 1.5 | - | |
2026.04 | 100 | 2.5 | - | |
2026.04 | 100 | 1 | - | |
2026.04 | 100 | 2.5 | - | |
2026.04 | 100 | 3.5 | - | |
2026.04 | 100 | 3.5 | - | |
2026.04 | 100 | 4 | - | |
2026.04 | 100 | 17 | - | |
2026.04 | 100 | 24.5 | - | |
2026.04 | 100 | 20 | - | |
2026.04 | 100 | 1.5 | - | |
2026.04 | 100 | 2 | - |