Share your thoughts, 1 month free Claude Pro on usSee more

Mathematical Reasoning on DART-Math (DM-1 to DM-5)

84.8DM-1 Score

FlexiDepth

Updated 1mo ago

Evaluation Results

Method	Links
FlexiDepth 2026.06		84.8	59.2	19.4	6.8	6.6
FlexiDepth 2026.06		81.2	55.6	18.4	5.2	4.8
FlexiDepth 2026.06		75.8	50.6	16.4	4.2	2.8
FlexiDepth 2026.06		67	42.2	13.8	3.2	2
POLAR 2026.06		55.6	44.8	25.4	18.8	17.8
POLAR 2026.06		53.4	44	24.8	18.4	17.6
POLAR 2026.06		50.8	41	22.2	18	17.2
FlexiDepth 2026.06		49.6	29.4	8.8	1.8	1
Base (sampling) 2026.06		48.4	27.8	13.6	13.4	10.6
POLAR 2026.06		48.4	33.2	21.2	16	14.6
Base (sampling) 2026.06		47	26.6	12.6	12.8	10.2
Base (sampling) 2026.06		45.8	26	12.4	11.4	9.8
POLAR 2026.06		43.2	30.4	20.8	13.5	14.9
Base (sampling) 2026.06		42	25.6	11.6	10.6	8.8
Base 2026.06		41.6	28.8	19	12	13
Base (sampling) 2026.06		37	23	9.2	9.6	7
DR.LLM 2026.06		2.4	0.6	0.8	0.2	2.4
DR.LLM 2026.06		2.4	0.6	0.8	0.2	2.4
DR.LLM 2026.06		2.4	0.6	0.8	0.2	2.8
DR.LLM 2026.06		2.4	1	0.8	0.2	2.8
DR.LLM 2026.06		2	0.6	0.6	0.2	2
ShortGPT 2026.06		0	0	0	0	0
ShortGPT 2026.06		0	0	0	0	0
ShortGPT 2026.06		0	0	0	0	0
ShortGPT 2026.06		0	0	0	0	0
ShortGPT 2026.06		0	0	0	0	0
MindSkip 2026.06		0	0	0	0	0
MindSkip 2026.06		0	0	0	0	0
MindSkip 2026.06		0	0	0	0	0
MindSkip 2026.06		0	0	0	0	0
MindSkip 2026.06		0	0	0	0	0