Prompt Injection Detection on HN-Inj. OOD
80.82Macro-F1TF–IDF only
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| TF–IDF onlyRegime=HN-Inj. OOD2026.05 | 80.82 | 81.12 | 90.79 | 92.32 | 23.2 | 54.47 | 75.22 | |
| DeBERTa-baseRegime=HN-Inj. OOD2026.05 | 73.13 | 74.73 | 86.41 | 87.84 | 35.2 | 59.98 | 64.31 | |
| LLM Guard PromptInjectionRegime=HN-Inj. OOD2026.05 | 45.37 | 45.43 | 42.28 | 43.41 | 0 | 0.29 | 0.88 |