List your company

Gender on Patchscopes few-shot 1.0 (test)

82.08SRIE
-2.014419.817841.6563.4822Jan 30, 2026

Evaluation results

MethodLinks
IEModel=Llama2-7b, Evaluation Protocol=few-shot2026.0182.08
DBModel=Llama2-7b, Evaluation Protocol=few-shot2026.0180.19
CBModel=Llama2-7b, Evaluation Protocol=few-shot2026.0169.34
BALOR (S)Model=Qwen3-4b, Evaluation Protocol=few-shot2026.0141.3
BALOR (D)Model=Llama2-7b, Evaluation Protocol=few-shot2026.0140.57
BALOR (D)Model=Qwen3-4b, Evaluation Protocol=few-shot2026.0139.13
BALOR (S)Model=Llama2-7b, Evaluation Protocol=few-shot2026.0137.74
FairSteerModel=Llama2-7b, Evaluation Protocol=few-shot2026.0136.8
BALOR (D)Model=Llama3.2-1b, Evaluation Protocol=few-shot2026.0136.67
Fine-tuningModel=Llama2-7b, Evaluation Protocol=few-shot2026.0136.23
VanillaModel=Llama2-7b, Evaluation Protocol=few-shot2026.0134.91
IEModel=Llama3-8b, Evaluation Protocol=few-shot2026.0134.15
BALOR (S)Model=Llama3-8b, Evaluation Protocol=few-shot2026.0129.27
CBModel=Llama3-8b, Evaluation Protocol=few-shot2026.0128.05
BALOR (D)Model=Llama3-8b, Evaluation Protocol=few-shot2026.0128.05
BALOR (S)Model=Llama3.2-1b, Evaluation Protocol=few-shot2026.0126.67
VanillaModel=Qwen3-4b, Evaluation Protocol=few-shot2026.0126.08
FairSteerModel=Llama3-8b, Evaluation Protocol=few-shot2026.0121.95
DBModel=Qwen3-4b, Evaluation Protocol=few-shot2026.0121.74
FairSteerModel=Qwen3-4b, Evaluation Protocol=few-shot2026.0121.74
CBModel=Qwen3-4b, Evaluation Protocol=few-shot2026.0117.39
Fine-tuningModel=Qwen3-4b, Evaluation Protocol=few-shot2026.0117.39
DBModel=Llama3-8b, Evaluation Protocol=few-shot2026.0117.07
CBModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.0116.67
IEModel=Qwen3-4b, Evaluation Protocol=few-shot2026.0115.21
DBModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.0113.33
Fine-tuningModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.0113.33
VanillaModel=Llama3-8b, Evaluation Protocol=few-shot2026.0112.2
IEModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.0110
FairSteerModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.0110
VanillaModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.016.67
Fine-tuningModel=Llama3-8b, Evaluation Protocol=few-shot2026.011.22