Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on GSM8K (Spearman Rho Correlation & Resolution Gain)
Loading...
1
Spearman Rho (Agreement)
FullPair
-0.6328
-0.2089
0.215
0.6389
May 28, 2026
Spearman Rho (Agreement)
Resolution Gain
Updated 1mo ago
Evaluation Results
Method
Method
Links
Spearman Rho (Agreement)
Resolution Gain
FullPair
Protocol=FullPair
2026.05
1
30.07
SEAL
Protocol=SEAL
2026.05
0.83
10.5
Pointwise
Protocol=Pointwise
2026.05
0.81
5.29
Listwise
Protocol=Listwise
2026.05
0.81
28.64
FlatBrk
Protocol=FlatBrk
2026.05
0.57
9.87
FixedRub
Protocol=FixedRub
2026.05
0.26
5.72
Original
Protocol=Original
2026.05
-0.57
1
Feedback
Search any
task
Search any
task
