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

Math Reasoning on AIME

51.67Accuracy

VCORE

Updated 28d ago

Evaluation Results

Method	Links
VCORE 2025.10		51.67	51.67
VCORE 2025.10		48.33	48.33
SFT 2025.10		48.33	48.33
SFT 2025.10		46.67	46.67
iw-SFT 2025.10		45	45
VCORE 2025.10		45	45
Original 2025.10		43.33	43.33
Random 2025.10		43.33	43.33
Original 2025.10		43.33	43.33
DFT 2025.10		43.33	43.33
DFT 2025.10		41.67	41.67
iw-SFT 2025.10		40	40
iw-SFT 2025.10		40	40
Random 2025.10		40	40
Original 2025.10		38.33	38.33
Random 2025.10		38.33	38.33
SFT 2025.10		38.33	38.33
DFT 2025.10		35	35
VCORE 2025.10		6.67	6.67
iw-SFT 2025.10		5	5
Original 2025.10		3.33	3.33
SFT 2025.10		3.33	3.33
DFT 2025.10		3.33	3.33
Random 2025.10		1.67	1.67