Safety Evaluation on HarmBench Std
80.8Refusal RateALIGNBEAM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ALIGNBEAMDraft model=Qwen2-7B-Base, Domain=same-vocab ctrl, LBM-DROP=true, alpha=0.5, N=6, K=3, seed=422026.06 | 80.8 | 89.8 | |
| ALIGNBEAMDraft model=Finance-Llama3-8B, Domain=finance, LBM-DROP=true, alpha=0.5, N=6, K=3, seed=422026.06 | 79.6 | 91 | |
| ALIGNBEAMDraft model=Llama-3.1-8B, Domain=general, LBM-DROP=true, alpha=0.5, N=6, K=3, seed=422026.06 | 79.6 | 91.6 | |
| ALIGNBEAMDraft model=MedLlama-3-8B, Domain=medical, LBM-DROP=true, alpha=0.5, N=6, K=3, seed=422026.06 | 78.4 | 88.6 | |
| ALIGNBEAMDraft model=Macro average, LBM-DROP=true, alpha=0.5, K=3, seed=422026.06 | 76.4 | 89.2 | |
| ALIGNBEAMDraft model=DeepSeek-Math-7B-Instruct, Domain=math, LBM-DROP=true, alpha=0.5, N=20, K=3, seed=422026.06 | 63.5 | 85 |