List your company

Instruction Following and Safety Evaluation on Alpaca

52.2BRT ScoreSFT
44.446.42548.4550.475May 22, 2025

Evaluation results

MethodLinks
SFTModel=Vicuna-7B2025.0552.24.9540.5
SAPModel=Llama2-7B2025.0552.16.0322.6
SAPModel=Vicuna-7B2025.0551.94.8724.9
SAPModel=Average2025.0551.95.5423.07
SafeInstrModel=Llama2-7B2025.0551.86.0628.9
SAPModel=Qwen2.5-7B2025.0551.65.7221.7
SFTModel=Average2025.0551.65.5537.43
SFTModel=Llama2-7B2025.0551.46.0633.1
SFTModel=Qwen2.5-7B2025.0551.25.6538.7
SafeInstrModel=Average2025.0551.15.5826.2
SafeInstrModel=Vicuna-7B2025.05514.9627.2
SaLoRAModel=Llama2-7B2025.0550.86.1529.2
BoosterModel=Vicuna-7B2025.0550.84.9928.4
LisaModel=Vicuna-7B2025.0550.65.2728.1
SafeInstrModel=Qwen2.5-7B2025.0550.45.7122.5
SAFTModel=Vicuna-7B2025.0550.35.0734.6
SaLoRAModel=Average2025.0550.35.7130.57
BoosterModel=Qwen2.5-7B2025.0550.25.7726.5
SaLoRAModel=Qwen2.5-7B2025.0550.25.8830.8
BoosterModel=Average2025.0550.15.6227.6
LisaModel=Average2025.0550.15.7625.93
LisaModel=Llama2-7B2025.0549.96.1725.4
SaLoRAModel=Vicuna-7B2025.0549.95.1131.7
LisaModel=Qwen2.5-7B2025.0549.85.8224.3
SAFTModel=Qwen2.5-7B2025.0549.65.7935.2
SAFTModel=Average2025.0549.55.6733.63
BoosterModel=Llama2-7B2025.0549.46.0927.9
VaccineModel=Vicuna-7B2025.0549.25.0330.4
VaccineModel=Qwen2.5-7B2025.0548.95.8128.8
VaccineModel=Average2025.0548.95.6629.23
VaccineModel=Llama2-7B2025.0548.76.1328.5
SAFTModel=Llama2-7B2025.0548.76.1431.1
Base modelModel=Vicuna-7B2025.0546.516.3832.3
Base modelModel=Qwen2.5-7B2025.0545.715.3434.3
Base modelModel=Average2025.0545.61732.5
Base modelModel=Llama2-7B2025.0544.719.2830.9