Hate Detection on Hate Detection
100ASRObliInjection
Evaluation Results
| Method | Links | |
|---|---|---|
| ObliInjectionVictim LLM=Llama-3-8B2025.12 | 100 | |
| ObliInjectionVictim LLM=Falcon3-7B2025.12 | 99.8 | |
| ObliInjectionVictim LLM=Qwen3-4B2025.12 | 99.5 | |
| ObliInjectionVictim LLM=Llama-3.1-8B2025.12 | 99 | |
| ObliInjectionVictim LLM=Average2025.12 | 97.3 | |
| ObliInjectionVictim LLM=Llama4-17B2025.12 | 96 | |
| ObliInjectionVictim LLM=Qwen-2.5-7B2025.12 | 95.8 | |
| ObliInjectionVictim LLM=Mistral-7B2025.12 | 91 |