Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Language Understanding on MMLU (5-shot Accuracy)
Loading...
78.82
5-shot Accuracy
LP-SFT
54.1616
60.5633
66.965
73.3667
Jul 6, 2026
5-shot Accuracy
Updated 18d ago
Evaluation Results
Method
Method
Links
5-shot Accuracy
LP-SFT
Backbone=Qwen3-14B, Fe...
2026.07
78.82
ASFT
Backbone=Qwen3-14B, Fe...
2026.07
77.7
EAFT
Backbone=Qwen3-14B, Fe...
2026.07
77.13
CE
Backbone=Qwen3-14B, Fe...
2026.07
76.99
GEM
Backbone=Qwen3-14B, Fe...
2026.07
76.96
DFT
Backbone=Qwen3-14B, Fe...
2026.07
76.11
LP-SFT
Backbone=Qwen3-4B, Few...
2026.07
70.32
ASFT
Backbone=Qwen3-4B, Few...
2026.07
69.54
EAFT
Backbone=Qwen3-4B, Few...
2026.07
69.08
GEM
Backbone=Qwen3-4B, Few...
2026.07
68.99
CE
Backbone=Qwen3-4B, Few...
2026.07
68.9
DFT
Backbone=Qwen3-4B, Few...
2026.07
66.82
LP-SFT
Backbone=Llama3.1-8B,...
2026.07
62.48
EAFT
Backbone=Llama3.1-8B,...
2026.07
61.14
CE
Backbone=Llama3.1-8B,...
2026.07
61.11
GEM
Backbone=Llama3.1-8B,...
2026.07
61
ASFT
Backbone=Llama3.1-8B,...
2026.07
60.58
DFT
Backbone=Llama3.1-8B,...
2026.07
55.11
Feedback
Search any
task
Search any
task