List your company

Open-ended Generation on TruthfulQA Without Rejected Samples open-ended (full)

74.67TruthfulnessCoCoASIG
48.139655.027361.91568.8027Feb 10, 2026

Evaluation results

MethodLinks
CoCoASIGModel=Llama-3-8b, Mode=ConMLDS (α=6.0)2026.0274.6780.660.18
CoCoASIGModel=Llama-3-8b, Mode=ConMLDS (α=2.5)2026.0274.3481.9160.89
CoCoASIGModel=Llama-3-8b, Mode=ConMLDS (α=1.0)2026.0273.6882.5760.84
CoCoASIGModel=Llama-3-8b, Mode=fMLDS (α=2.5)2026.0273.2380.5859.01
CoCoASIGModel=Llama-3-8b, Mode=fMLDS (α=1.0)2026.0272.8879.4157.87
CoCoASIGModel=Llama-3-8b, Mode=fMLDS (α=6.0)2026.0272.6780.3958.42
DiverModel=Qwen-2.5-32b2026.0269.9479.7855.8
CoCoASIGModel=Mistral-7b, Mode=fMLDS (α=2.5)2026.0268.6482.8456.86
CoCoASIGModel=Qwen-2.5-32b, Mode=ConMLDS (α=1.0)2026.0268.4984.8958.14
CoCoASIGModel=Qwen-2.5-32b, Mode=fMLDS (α=1.0)2026.0268.1786.558.96
CoCoASIGModel=Qwen-2.5-32b, Mode=fMLDS (α=2.5)2026.0268.1787.7859.84
CoCoASIGModel=Mistral-7b, Mode=fMLDS (α=6.0)2026.0268.1481.9555.84
CoCoASIGModel=Mistral-7b, Mode=ConMLDS (α=2.5)2026.0267.9583.7356.9
CoCoASIGModel=Qwen-2.5-32b, Mode=ConMLDS (α=2.5)2026.0267.8588.7560.21
CoCoASIGModel=Mistral-7b, Mode=ConMLDS (α=6.0)2026.0267.6682.4955.81
CoCoASIGModel=Qwen-2.5-32b, Mode=fMLDS (α=6.0)2026.0267.5284.2456.89
CoCoASIGModel=Mistral-7b, Mode=fMLDS (α=1.0)2026.0267.3683.0955.97
CoCoASIGModel=Qwen-2.5-32b, Mode=ConMLDS (α=6.0)2026.0267.285.8557.69
CoCoASIGModel=Mistral-7b, Mode=ConMLDS (α=1.0)2026.0267.1684.6256.83
DoLaModel=Llama-3-8b2026.0266.9773.6849.34
BaselineModel=Qwen-2.5-32b2026.0266.4688.2758.67
BaselineModel=Mistral-7b2026.0266.1483.9355.51
DoLaModel=Mistral-7b2026.0264.9883.3854.19
DoLaModel=Qwen-2.5-32b2026.0263.7276.1148.49
DiverModel=Mistral-7b2026.0263.5681.6251.87
DiverModel=Llama-3-8b2026.0263.3570.9344.93
DoLaModel=Qwen-2.5-14b2026.0262.0972.5345.03
DiverModel=Qwen-2.5-14b2026.0261.4777.5147.65
BaselineModel=Llama-3-8b2026.0260.6965.9240.01
BaselineModel=Qwen-2.5-14b2026.026078.947.34
CoCoASIGModel=Qwen-2.5-14b, Mode=ConMLDS (α=2.5)2026.0259.3686.0651.08
CoCoASIGModel=Qwen-2.5-14b, Mode=ConMLDS (α=1.0)2026.0259.0484.3449.79
CoCoASIGModel=Qwen-2.5-14b, Mode=fMLDS (α=2.5)2026.0258.9485.3750.32
CoCoASIGModel=Qwen-2.5-14b, Mode=ConMLDS (α=6.0)2026.0258.3383.3348.61
CoCoASIGModel=Qwen-2.5-14b, Mode=fMLDS (α=1.0)2026.0258.1385.7749.86
CoCoASIGModel=Qwen-2.5-14b, Mode=fMLDS (α=6.0)2026.0258.1384.1548.91
DeCoReModel=Mistral-7b2026.0255.3483.2146.05
DeCoReModel=Llama-3-8b2026.0255.0987.1748.03
DeCoReModel=Qwen-2.5-14b2026.0249.1647.4923.35