Color on Patchscopes few-shot 1.0 (test)
69.77SRBALOR (S)
Evaluation Results
| Method | Links | |
|---|---|---|
| BALOR (S)Model=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 69.77 | |
| BALOR (S)Model=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 68.9 | |
| BALOR (D)Model=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 67.54 | |
| BALOR (D)Model=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 65.99 | |
| BALOR (S)Model=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 52.1 | |
| BALOR (D)Model=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 48.19 | |
| BALOR (S)Model=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 47.62 | |
| BALOR (D)Model=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 43.47 | |
| Fine-tuningModel=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 41.72 | |
| Fine-tuningModel=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 29.17 | |
| FairSteerModel=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 22.09 | |
| CBModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 21.07 | |
| FairSteerModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 19.77 | |
| VanillaModel=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 18.41 | |
| Fine-tuningModel=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 18.37 | |
| DBModel=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 15.5 | |
| DBModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 15.12 | |
| FairSteerModel=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 14.5 | |
| IEModel=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 14.44 | |
| VanillaModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 14.24 | |
| CBModel=Qwen3-4b, Evaluation Protocol=few-shot2026.01 | 13.66 | |
| VanillaModel=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 10.97 | |
| IEModel=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 8.7 | |
| DBModel=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 8.6 | |
| IEModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 8.14 | |
| CBModel=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 7.83 | |
| CBModel=Llama3-8b, Evaluation Protocol=few-shot2026.01 | 6.63 | |
| VanillaModel=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 5.42 | |
| FairSteerModel=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 4.82 | |
| Fine-tuningModel=Llama3.2-1b, Evaluation Protocol=few-shot2026.01 | 3.78 | |
| IEModel=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 3.01 | |
| DBModel=Llama2-7b, Evaluation Protocol=few-shot2026.01 | 1.51 |