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

Reasoning on Multi-Step Arithmetic

99.2Accuracy

RoT

33.57650.61367.6584.687Oct 16, 2024Jan 6, 2025Mar 29, 2025Jun 19, 2025Sep 9, 2025Nov 30, 2025Feb 20, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2024.10
99.2-
2024.10
98.4-
2024.10
96.8-
2024.10
94.8-
2024.10
90.7-
2024.10
90.3-
2024.10
89.6-
2024.10
89.5-
2024.10
89.2-
2024.10
88.8-
2024.10
88.7-
2024.10
88.3-
2024.10
87.6-
2024.10
87.6-
2024.10
85.5-
2024.10
85.1-
2024.10
83.1-
2024.10
83.1-
2024.10
82.3-
2024.10
81.9-
2024.10
79.1-
2024.10
79-
2024.10
76.2-
2024.10
73.9-
2026.02
45.815.259
2026.02
45.614.5
2026.02
43.684.609
2026.02
36.12.516