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

Arithmetic Reasoning on MultiArith (test)

99.5Accuracy

LoRAALL

58.9469.478090.53Oct 28, 2022Jun 4, 2023Jan 9, 2024Aug 15, 2024Mar 22, 2025Oct 27, 2025Jun 4, 2026
Updated 29d ago

Evaluation Results

MethodLinks
2024.10
99.5
2024.10
99.5
2023.11
99.3
2023.11
99.2
2024.05
99
2023.11
99
2024.10
99
2026.05
98.87
2026.04
98.8
2024.10
98.8
2024.10
98.8
2024.10
98.8
2026.05
98.72
2026.04
98.54
2026.05
98.51
2024.05
98.5
2026.05
98.5
2026.05
98.5
2026.05
98.45
2026.05
98.43
2026.05
98.42
2026.05
98.23
2026.05
98.23
2026.05
98.22
2023.11
98.2
2023.11
98.2
2024.05
98.17
2026.05
98.12
2024.05
98
2026.05
97.81
2026.04
97.77
2026.05
97.54
2022.10
97.5
2026.04
97.5
2026.04
97.49
2026.05
97.42
2026.04
97.33
2026.04
97.29
2026.04
97.12
2026.05
97
2024.10
97
2026.04
96.85
2026.04
96.79
2026.04
96.7
2026.05
96.7
2026.04
96.58
2026.05
96.5
2026.04
96.42
2026.04
96.31
2026.04
96.22
2026.05
96.2
2026.02
96.1
2026.04
96.03
2026.04
96.02
2026.05
96
2026.05
96
2026.05
95.8
2026.05
95.3
2026.05
95.1
2026.05
94.9
2023.11
94.7
2026.05
94.7
2026.05
94.7
2026.05
94.7
2026.05
94.7
2026.05
94.7
2026.02
92.8
2022.10
92.3
2023.11
92.3
2023.11
88.5
2023.11
88.5
2026.06
88
2026.06
87.8
2023.11
87
2026.05
86.8
2023.11
86
2026.02
85
2026.06
84.8
2024.10
84.6
2024.10
84.5
2024.10
84.3
2024.10
84.2
2024.10
82.2
2023.11
82
2026.06
81.8
2024.10
81.4
2024.10
80.5
2023.11
78.7
2026.02
78.3
2022.10
77.3
2023.11
75.7
2026.02
74.4
2023.11
74.3
2023.11
73.5
2023.11
73.5
2023.11
71.5
2026.02
67.8
2026.02
64.4
2026.02
63.3
2022.10
60.5
Showing 100 of 136 rows