Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
CUBIC on Synthetic Arithmetic Tasks
Loading...
10,000
Success Rate
Global-Local Pipeline
-400
2,300
5,000
7,700
Jun 30, 2025
Success Rate
Updated 4d ago
Evaluation Results
Method
Method
Links
Success Rate
Global-Local Pipeline
Initialization=P_b^-,...
2025.06
10,000
Global-Local Pipeline
Initialization=P_s^-,...
2025.06
380
Global-Local Pipeline
Initialization=P_s^-,...
2025.06
190
Reverse Order
Target Length=L = 20
2025.06
160
Reverse Order
Target Length=L = 13
2025.06
140
Global-Local Pipeline
Initialization=P_b^-,...
2025.06
70
Reverse Order
Target Length=L = 10
2025.06
60
Global-Local Pipeline
Initialization=P_r^-,...
2025.06
30
Global-Local Pipeline
Initialization=P_r^-,...
2025.06
20
Global-Local Pipeline
Initialization=P_r^-,...
2025.06
0
Global-Local Pipeline
Initialization=P_s^-,...
2025.06
0
Global-Local Pipeline
Initialization=P_b^-,...
2025.06
0
Reverse Order
Target Length=L = 30
2025.06
0
Reverse Order
Target Length=L = 40
2025.06
0
Feedback
Search any
task
Search any
task