Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on GSM8K (ΔLL)
Loading...
28.8
Delta LL (GSM8K)
CE
-3.35056
4.99622
13.343
21.68978
May 28, 2026
Delta LL (GSM8K)
Updated 5d ago
Evaluation Results
Method
Method
Links
Delta LL (GSM8K)
CE
Backbone=7B, Adapter=DoRA
2026.05
28.8
CE
Backbone=7B, Adapter=LoRA
2026.05
28.5
CE
Backbone=7B, Adapter=S...
2026.05
27.9
CE
Backbone=0.5B, Adapter...
2026.05
22.4
CE
Backbone=7B, Adapter=R...
2026.05
22.4
CE
Backbone=0.5B, Adapter...
2026.05
22.1
CE
Backbone=0.5B, Adapter...
2026.05
21.5
CE
Backbone=0.5B, Adapter...
2026.05
17.3
CE + TMKL
Backbone=0.5B, Adapter...
2026.05
0.5
CE + TMKL
Backbone=0.5B, Adapter...
2026.05
0.4
CE + TMKL
Backbone=0.5B, Adapter...
2026.05
0.3
CE + TMKL
Backbone=7B, Adapter=DoRA
2026.05
0.3
CE + TMKL
Backbone=7B, Adapter=LoRA
2026.05
0.2
CE + TMKL
Backbone=7B, Adapter=S...
2026.05
0.1
CE + TMKL
Backbone=0.5B, Adapter...
2026.05
-0.1
CE + TMKL
Backbone=7B, Adapter=R...
2026.05
-0.4
Base
Backbone=7B
2026.05
-1.245
Base
Backbone=0.5B
2026.05
-2.114
Feedback
Search any
task
Search any
task