Multilingual Mathematical Reasoning on MT Math100

60.1Accuracy

SP3F-7B

Updated 4d ago

Evaluation Results

Method
SP3F-7B 2026.01	60.1	87	-
Qwen2.5-7B-Instruct + Translate Test 2026.01	60.08	-	59.34
SP3F-7B 2026.01	59.8	94.4	-
SP3F-7B 2026.01	59.1	83.7	-
Qwen2.5-7B-Instruct 2026.01	58.5	58.8	-
Qwen2.5-7B-Instruct 2026.01	56.9	75.4	-
SP3F-7B 2026.01	56.84	-	82.93
SP3F-7B 2026.01	56.4	75.5	-
Qwen2.5-7B-Instruct 2026.01	55.9	62.9	-
SP3F-7B 2026.01	53.3	92	-
Qwen2.5-7B-Instruct 2026.01	52.7	36.4	-
Qwen2.5-7B-Instruct 2026.01	52.12	-	65.66
SP3F-7B 2026.01	51.8	87.1	-
SP3F-7B 2026.01	51.7	83.6	-
Qwen2.5-7B-Instruct 2026.01	50.6	70.7	-
Qwen2.5-7B-Instruct 2026.01	48.3	62.7	-
Qwen2.5-7B + RLVR 2026.01	44.5	-	86.1
Qwen2.5-7B-Instruct 2026.01	44.4	56.7	-
SP3F-7B 2026.01	38	68.9	-
Qwen2.5-7B-Instruct 2026.01	35.5	65	-
SP3F-7B 2026.01	35.1	80.3	-
Qwen2.5-7B-Instruct 2026.01	32.1	75.5	-
Qwen2.5-7B + SFT 2026.01	26.72	-	58.26
Qwen2.5-7B 2026.01	21.16	-	58.22