List your company

Mathematical Reasoning on MATH (test) (Accuracy, Reasoning Length)

17.31AccuracyLTF
4.69487.969911.24514.5201Jun 15, 2026

Evaluation results

MethodLinks
LTFBackbone=LLaMA-8B, Finetuning setting=True, Extreme compression=True2026.0617.311
LTFBackbone=DS-1.5B, Finetuning setting=True, Extreme compression=True2026.0617.21
LTFBackbone=LLaMA-3B, Finetuning setting=True, Extreme compression=True2026.0616.181
ReGuLaRBackbone=DS-1.5B, Finetuning setting=True, Extreme compression=True2026.0615.61
LTFBackbone=Average, Finetuning setting=True, Extreme compression=True2026.0614.71
ReGuLaRBackbone=LLaMA-8B, Finetuning setting=True, Extreme compression=True2026.0613.91
ReGuLaRBackbone=Average, Finetuning setting=True, Extreme compression=True2026.0611.981
ReGuLaRBackbone=LLaMA-3B, Finetuning setting=True, Extreme compression=True2026.0611.81
CoLaRBackbone=DS-1.5B, Finetuning setting=True, Extreme compression=True2026.0611.1362.2
CoLaRBackbone=LLaMA-8B, Finetuning setting=True, Extreme compression=True2026.069.1367.19
CoconutBackbone=LLaMA-8B, Finetuning setting=True, Extreme compression=True2026.068.686
CoLaRBackbone=Average, Finetuning setting=True, Extreme compression=True2026.068.4862.03
CoLaRBackbone=LLaMA-3B, Finetuning setting=True, Extreme compression=True2026.068.3560.44
LTFBackbone=LLaMA-1B, Finetuning setting=True, Extreme compression=True2026.068.11
CoconutBackbone=LLaMA-3B, Finetuning setting=True, Extreme compression=True2026.067.726
CoconutBackbone=Average, Finetuning setting=True, Extreme compression=True2026.067.066
CoconutBackbone=DS-1.5B, Finetuning setting=True, Extreme compression=True2026.066.676
ReGuLaRBackbone=LLaMA-1B, Finetuning setting=True, Extreme compression=True2026.066.621
CoLaRBackbone=LLaMA-1B, Finetuning setting=True, Extreme compression=True2026.065.3158.29
CoconutBackbone=LLaMA-1B, Finetuning setting=True, Extreme compression=True2026.065.186