List your company

Mathematical Reasoning on AIME 2024 (Accuracy, Length, CR)

73.3AccuracyDEER
14.43629.7184560.282Jun 13, 2026

Evaluation results

MethodLinks
DEERBackbone=Qwen3-14B2026.0673.38,40279.7
ASAGBackbone=Qwen3-14B2026.0673.37,60272.1
ASAGBackbone=Qwen3-4B2026.06708,76873.6
VanillaBackbone=Qwen3-14B2026.067010,537100
TALEBackbone=Qwen3-14B2026.067010,35898.3
DynasorBackbone=Qwen3-14B2026.067011,156105.9
ASAGBackbone=Qwen3-8B2026.0666.78,68371.7
VanillaBackbone=Qwen3-4B2026.0663.311,916100
DynasorBackbone=Qwen3-4B2026.0663.311,38795.6
VanillaBackbone=Qwen3-8B2026.0663.312,101100
DEERBackbone=Qwen3-8B2026.0663.38,93073.8
TALEBackbone=Qwen3-4B2026.066011,46996.2
DEERBackbone=Qwen3-4B2026.06608,74573.4
TALEBackbone=Qwen3-8B2026.066011,84797.9
DynasorBackbone=Qwen3-8B2026.066011,36593.9
ASAGBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0646.79,15466.7
DEERBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0643.38,66963.2
VanillaBackbone=DeepSeek-R1-Distill-Qwen-7B2026.064013,722100
DynasorBackbone=DeepSeek-R1-Distill-Qwen-7B2026.064011,46283.5
TALEBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0633.311,62784.7
NoThinkingBackbone=Qwen3-14B2026.06307,72673.3
NoThinkingBackbone=Qwen3-4B2026.0623.36,45854.2
NoThinkingBackbone=Qwen3-8B2026.0623.37,12158.8
NoThinkingBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0616.77,36553.7