List your company

Open ended generation on TruthfulQA With All Samples open-ended (full)

82.75TruthfulnessDoLa
64.0368.8973.7578.61Feb 10, 2026

Evaluation results

MethodLinks
DoLaModel=Qwen-2.5-14b2026.0282.7533.2554.527.51
CoCoASIGModel=Llama-3-8b, Mode=ConMLDS (α=6.0)2026.028160.422.8948.92
CoCoASIGModel=Llama-3-8b, Mode=ConMLDS (α=2.5)2026.0280.562.2522.1550.11
CoCoASIGModel=Llama-3-8b, Mode=ConMLDS (α=1.0)2026.028062.7522.0550.2
CoCoASIGModel=Llama-3-8b, Mode=fMLDS (α=1.0)2026.0279.2560.7521.4248.14
CoCoASIGModel=Llama-3-8b, Mode=fMLDS (α=2.5)2026.0279.2562.4120.6949.46
DoLaModel=Qwen-2.5-32b2026.0279.2543.2543.534.28
CoCoASIGModel=Llama-3-8b, Mode=fMLDS (α=6.0)2026.0278.7562.520.4449.22
CoCoASIGModel=Qwen-2.5-32b, Mode=ConMLDS (α=1.0)2026.0275.566.6722.0550.33
CoCoASIGModel=Qwen-2.5-32b, Mode=fMLDS (α=1.0)2026.0275.2567.592050.86
CoCoASIGModel=Qwen-2.5-32b, Mode=fMLDS (α=2.5)2026.0275.2568.5920.5151.62
CoCoASIGModel=Qwen-2.5-32b, Mode=ConMLDS (α=2.5)2026.027569.1718.4651.88
CoCoASIGModel=Qwen-2.5-14b, Mode=fMLDS (α=2.5)2026.0274.7552.538.539.24
CoCoASIGModel=Qwen-2.5-32b, Mode=fMLDS (α=6.0)2026.0274.7565.6620.2649.08
CoCoASIGModel=Qwen-2.5-14b, Mode=ConMLDS (α=1.0)2026.0274.552.537.7539.11
CoCoASIGModel=Qwen-2.5-14b, Mode=ConMLDS (α=2.5)2026.0274.55437.2540.23
CoCoASIGModel=Qwen-2.5-32b, Mode=ConMLDS (α=6.0)2026.0274.566.7521.2849.73
CoCoASIGModel=Qwen-2.5-14b, Mode=fMLDS (α=1.0)2026.0274.2552.7538.539.17
CoCoASIGModel=Qwen-2.5-14b, Mode=fMLDS (α=6.0)2026.0274.2551.7538.538.42
CoCoASIGModel=Qwen-2.5-14b, Mode=ConMLDS (α=6.0)2026.0273.7552.53738.72
DiverModel=Qwen-2.5-32b2026.027371.331152.07
BaselineModel=Mistral-7b2026.0272.7568.5920.2549.9
BaselineModel=Qwen-2.5-32b2026.0272.7571.6818.7552.15
DoLaModel=Mistral-7b2026.0272.2567.420.7548.7
CoCoASIGModel=Mistral-7b, Mode=fMLDS (α=2.5)2026.0272.2573.2515.552.92
CoCoASIGModel=Mistral-7b, Mode=ConMLDS (α=2.5)2026.0271.9374.2515.553.41
DoLaModel=Llama-3-8b2026.0271.7561.4616.7544.1
CoCoASIGModel=Mistral-7b, Mode=fMLDS (α=6.0)2026.0271.7572.9315.2552.33
CoCoASIGModel=Mistral-7b, Mode=ConMLDS (α=6.0)2026.0271.7572.515.7552.02
CoCoASIGModel=Mistral-7b, Mode=fMLDS (α=1.0)2026.0271.2573.2515.7552.19
CoCoASIGModel=Mistral-7b, Mode=ConMLDS (α=1.0)2026.0271.2574.7515.553.26
DeCoReModel=Qwen-2.5-14b2026.0271.2527.9655.2519.92
DeCoReModel=Mistral-7b2026.0269.558.7534.540.83
BaselineModel=Qwen-2.5-14b2026.026961.8722.542.69
DeCoReModel=Llama-3-8b2026.0268.57133.7548.63
DiverModel=Mistral-7b2026.0268.572.614.2549.73
BaselineModel=Llama-3-8b2026.026657.2813.537.81
DiverModel=Qwen-2.5-14b2026.0265.7569.2311.7545.52
DiverModel=Llama-3-8b2026.0264.7567.874.543.94