List your company

Mathematical Reasoning on GSM8K (Accuracy, Steps, Speed)

82.56AccuracyCOVER
69.102472.596276.0979.5838Feb 5, 2026

Evaluation results

MethodLinks
COVERBackbone=LLaDA-1.5-8B, Length=5122026.0282.5663.845.41
baselineBackbone=LLaDA-1.5-8B, Length=5122026.0282.345121
baselineBackbone=LLaDA-1.5-8B, Length=2562026.0282.112561
WINOBackbone=LLaDA-1.5-8B, Length=5122026.0282.0374.733.72
COVERBackbone=LLaDA-1.5-8B, Length=2562026.0281.7363.242.65
SaberBackbone=LLaDA-1.5-8B, Length=2562026.0281.43107.322.33
SaberBackbone=LLaDA-1.5-8B, Length=5122026.0281.35117.534.29
WINOBackbone=LLaDA-1.5-8B, Length=2562026.0281.1285.032.44
COVERBackbone=LLaDA-Ins-8B, Length=5122026.0279.9865.475.52
SaberBackbone=LLaDA-Ins-8B, Length=5122026.0279.61128.753.87
WINOBackbone=LLaDA-Ins-8B, Length=5122026.0279.4574.963.61
COVERBackbone=Dream-Ins-7B, Length=5122026.0279.351.488.48
baselineBackbone=LLaDA-Ins-8B, Length=5122026.0279.085121
COVERBackbone=Dream-Ins-7B, Length=2562026.0278.9252.113.51
WINOBackbone=LLaDA-Ins-8B, Length=2562026.0277.3349.473.01
COVERBackbone=LLaDA-Ins-8B, Length=2562026.0277.2651.653.25
baselineBackbone=Dream-Ins-7B, Length=5122026.0276.885121
baselineBackbone=Dream-Ins-7B, Length=2562026.0275.822561
SaberBackbone=LLaDA-Ins-8B, Length=2562026.0275.66120.482.47
baselineBackbone=LLaDA-Ins-8B, Length=2562026.0274.912561
WINOBackbone=Dream-Ins-7B, Length=5122026.0274.0796.985.05
WINOBackbone=Dream-Ins-7B, Length=2562026.0274833.01
SaberBackbone=Dream-Ins-7B, Length=5122026.0272.1450.31.26
COVERBackbone=LLaDA-Base-8B, Length=5122026.0271.42105.153.17
SaberBackbone=Dream-Ins-7B, Length=2562026.0271.04203.551.34
COVERBackbone=LLaDA-Base-8B, Length=2562026.0270.9697.211.78
baselineBackbone=LLaDA-Base-8B, Length=5122026.0270.745121
baselineBackbone=LLaDA-Base-8B, Length=2562026.0270.582561
WINOBackbone=LLaDA-Base-8B, Length=2562026.0270.43106.761.09
SaberBackbone=LLaDA-Base-8B, Length=5122026.0270.13166.083.07
WINOBackbone=LLaDA-Base-8B, Length=5122026.0270.05128.31.49
SaberBackbone=LLaDA-Base-8B, Length=2562026.0269.62135.361.72