Vulnerability scanning and safety evaluation on ANSIRAW
-566.3ELPDBAYES (S=1)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| BAYES (S=1)LLM=Mamba-2.8B2025.10 | -566.3 | 45.5 | 31 | |
| FREQ (naive)LLM=Mamba-2.8B2025.10 | -565.6 | 45.5 | 31 | |
| BAYES (S=n)LLM=Mamba-2.8B2025.10 | -558.1 | 45.5 | 32 | |
| BAYES (S=1)LLM=Pythia-2.8B2025.10 | -427.5 | 33 | 20 | |
| FREQ (naive)LLM=Pythia-2.8B2025.10 | -427 | 33 | 21 | |
| BAYES (S=n)LLM=Pythia-2.8B2025.10 | -426.4 | 33.7 | 23 | |
| BHM-ESC-MiniLLM=Pythia-2.8B2025.10 | -203.1 | 19.6 | 24 | |
| ORACLELLM=Pythia-2.8B2025.10 | -195.1 | 21.4 | 23 | |
| BHM-ESC-TFLLM=Pythia-2.8B2025.10 | -142.4 | 15.3 | 24 | |
| BHM-ESC-MiniLLM=Mamba-2.8B2025.10 | -132.8 | 12.2 | 26 | |
| ORACLELLM=Mamba-2.8B2025.10 | -115.2 | 12.9 | 20 | |
| BHM-ESC-TFLLM=Mamba-2.8B2025.10 | -97.2 | 10.7 | 25 |