List your company

Violated Principle Classification on ETHICSCAFF

94.55AccuracyQwen3-14B
15.062835.698956.33576.9711Aug 12, 2025

Evaluation results

MethodLinks
Qwen3-14BModel Family=Qwen3-14B, Variant=w/ ETHICSCAFF (Full)2025.0894.5595.2194.5594.37
Qwen3-8BModel Family=Qwen3-8B, Variant=w/ ETHICSCAFF (Full)2025.0888.9891.1188.9888.63
Qwen3-14BModel Family=Qwen3-14B, Variant=w/ ETHICSCAFF (SFT)2025.0874.3375.2774.3370.39
GPT-4oModel Family=Closed-Source Models, Variant=RAG2025.0873.6181.5273.6173.12
Qwen3-8BModel Family=Qwen3-8B, Variant=w/ ETHICSCAFF (SFT)2025.0873.6182.1773.6175.2
Qwen3-14BModel Family=Qwen3-14B, Variant=RAG2025.0868.4877.6968.4868.4
Llama3.1-8BModel Family=Llama3.1-8B, Variant=w/ ETHICSCAFF2025.0867.9663.2267.9662.22
Qwen3-14BModel Family=Qwen3-14B, Variant=Few-shot + CoT2025.0864.1570.9964.1559.87
Qwen3-8BModel Family=Qwen3-8B, Variant=Few-shot + CoT2025.0857.6258.9557.6251.55
Llama3.1-8BModel Family=Llama3.1-8B, Variant=RAG2025.0857.2262.2757.2254.98
Qwen3-8BModel Family=Qwen3-8B, Variant=RAG2025.0855.0263.5155.0252.02
Claude-Sonnet-4Model Family=Closed-Source Models, Variant=Base2025.0854.5965.8254.5947.44
Qwen3-14BModel Family=Qwen3-14B, Variant=Few-shot2025.0852.2461.5952.2447.04
DeepSeek-V3-685BModel Family=Open-Source Models, Variant=Base2025.0852.1464.0452.1446.85
GPT-4oModel Family=Closed-Source Models, Variant=Base2025.0851.1862.6951.1845.66
Qwen3-14BModel Family=Qwen3-14B, Variant=CoT2025.0850.9459.3450.9446.92
Qwen3-8BModel Family=Qwen3-8B, Variant=CoT2025.0850.2257.0350.2244.8
Qwen3-14BModel Family=Qwen3-14B, Variant=Base2025.0842.3166.3542.3138.67
Qwen3-8BModel Family=Qwen3-8B, Variant=Few-shot2025.0833.5546.1333.5531.58
Llama3.1-8BModel Family=Llama3.1-8B, Variant=Base2025.0829.3943.7329.3925.96
Qwen3-8BModel Family=Qwen3-8B, Variant=Base2025.0823.3344.1623.3322.18
Llama3.1-8BModel Family=Llama3.1-8B, Variant=Few-shot2025.0818.122218.1216.07