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

Mathematical Reasoning on AQuA (Exact Match %)

79.92AQuA Exact Match

Min-k

Updated 4d ago

Evaluation Results

Method	Links
Min-k 2026.04		79.92
Min-k 2026.04		79.13
Min-k 2026.04		79.13
Min-k 2026.04		79.13
Min-k 2026.04		79.13
Top-nσ 2026.04		78.35
Top-nσ 2026.04		78.35
Min-k 2026.04		78.35
Top-p 2026.04		77.95
Top-k 2026.04		77.56
Min-p 2026.04		77.56
Top-p 2026.04		77.17
Top-nσ 2026.04		77.17
Top-nσ 2026.04		77.17
Top-nσ 2026.04		77.17
Top-nσ 2026.04		77.17
Min-p 2026.04		76.77
Top-k 2026.04		75.2
Min-p 2026.04		74.41
Top-k 2026.04		71.65
Top-p 2026.04		70.87
Min-p 2026.04		69.69
Top-k 2026.04		62.2
Min-p 2026.04		55.12
Min-k 2026.04		52.76
Top-p 2026.04		52.76
Min-k 2026.04		50.39
Min-p 2026.04		50
Min-k 2026.04		50
Top-p 2026.04		49.61
Top-nσ 2026.04		49.61
Top-nσ 2026.04		48.03
Min-k 2026.04		48.03
Top-nσ 2026.04		47.24
Top-k 2026.04		46.85
Top-nσ 2026.04		46.06
Top-nσ 2026.04		46.06
Min-k 2026.04		46.06
Top-nσ 2026.04		44.49
Min-k 2026.04		44.49
Top-k 2026.04		44.09
Top-k 2026.04		40.94
Min-p 2026.04		32.68
Top-p 2026.04		31.1
Min-p 2026.04		24.02
Top-p 2026.04		19.29
Top-k 2026.04		17.72
Top-p 2026.04		12.99
Top-k 2026.04		8.66
Min-p 2026.04		7.48
Min-p 2026.04		7.08
Top-k 2026.04		6.69
Top-p 2026.04		3.54
Top-p 2026.04		1.18
Top-k 2026.04		0
Top-p 2026.04		0
Min-p 2026.04		0
Top-k 2026.04		0
Top-p 2026.04		0
Min-p 2026.04		0