List your company

Off-Topic Evaluation on OffTopicEval

98.89AR^IDLlama-3.1-8B +CD
22.1942.102562.01581.9275Jan 10, 2026

Evaluation results

MethodLinks
Llama-3.1-8B +CDBackbone=Llama-3.1-8B, Method=CD, alpha=02026.0198.8917.931.3717.58
Qwen-2.5-32B +CDBackbone=Qwen-2.5-32B, Method=CD, alpha=02026.0190.4351.8917.6850.24
Qwen-2.5-14B +CDBackbone=Qwen-2.5-14B, Method=CD, alpha=02026.0188.8233.0744.6954.08
Qwen-2.5-14B +CDBackbone=Qwen-2.5-14B, Method=CD, alpha=12026.0183.7846.4556.5263.78
Olmo-3.1-32B +CDBackbone=Olmo-3.1-32B, Method=CD, alpha=02026.0180.3150.2913.5245.67
Qwen-2.5-32B +CDBackbone=Qwen-2.5-32B, Method=CD, alpha=12026.0179.9261.2936.9260.83
Qwen-2.5-14B +CDBackbone=Qwen-2.5-14B, Method=CD, alpha=22026.0175.0359.1867.2768.62
Qwen-2.5-32B +CDBackbone=Qwen-2.5-32B, Method=CD, alpha=22026.0173.1278.5643.1666.43
Llama-3.1-8B +CDBackbone=Llama-3.1-8B, Method=CD, alpha=12026.0172.5460.6315.3749.87
Olmo-3.1-32B +CDBackbone=Olmo-3.1-32B, Method=CD, alpha=12026.0165.9675.522.4656.22
Olmo-3.1-32B +CDBackbone=Olmo-3.1-32B, Method=CD, alpha=22026.0161.0377.2129.9257.05
Qwen-2.5-7B +CDBackbone=Qwen-2.5-7B, Method=CD, alpha=02026.0153.3165.535.1451.77
Llama-3.1-8B +CDBackbone=Llama-3.1-8B, Method=CD, alpha=22026.0143.8980.7130.3549.03
Qwen-2.5-7B +CDBackbone=Qwen-2.5-7B, Method=CD, alpha=12026.0136.4876.1743.7545.36
Qwen-2.5-7B +CDBackbone=Qwen-2.5-7B, Method=CD, alpha=22026.0125.1485.9862.1937.54