Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on MATH (Accuracy, Delta)
Loading...
76.12
Accuracy
Qwen2.5(14B)T
23.0384
36.8192
50.6
64.3808
Jul 4, 2026
Accuracy
Delta
Updated 18d ago
Evaluation Results
Method
Method
Links
Accuracy
Delta
Qwen2.5(14B)T
Generation Length=8096
2026.07
76.12
-
RG-OPD (ours)
Generation Length=8096
2026.07
52.6
11.51
Reverse KL [1]
Generation Length=8096
2026.07
49.8
8.71
Qwen2.5(1.5B)S
Generation Length=8096
2026.07
41.09
0
Qwen2.5(14B)T
Generation Length=1024
2026.07
30.33
-
RG-OPD (ours)
Generation Length=1024
2026.07
28.63
1.16
TSD-KD [10]
Generation Length=8096
2026.07
27.89
-13.2
Qwen2.5(1.5B)S
Generation Length=1024
2026.07
27.47
0
TSD-KD [10]
Generation Length=1024
2026.07
27.1
-0.37
TSD-KD [10]*
Generation Length=1024
2026.07
26.1
-1.37
Reverse KL [1]
Generation Length=1024
2026.07
25.08
-2.39
Feedback
Search any
task
Search any
task