Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Commonsense Reasoning on HellaSwag non-IID distribution (alpha=0.1)
Loading...
58.47
Accuracy
FedAlign-MoE
50.7532
52.7566
54.76
56.7634
Mar 22, 2026
Accuracy
Updated 25d ago
Evaluation Results
Method
Method
Links
Accuracy
FedAlign-MoE
Model=Switch-base-16
2026.03
58.47
FedAlign-MoE
Model=DeepSeek-MoE-16B
2026.03
57.11
FedMoE
Model=Switch-base-16
2026.03
56.42
FedMoE
Model=DeepSeek-MoE-16B
2026.03
54.8
PFL-MoE
Model=Switch-base-16
2026.03
54.76
PFL-MoE
Model=DeepSeek-MoE-16B
2026.03
53.44
FedProx
Model=Switch-base-16
2026.03
52.88
FedAvg
Model=Switch-base-16
2026.03
52.06
FedProx
Model=DeepSeek-MoE-16B
2026.03
51.79
FedAvg
Model=DeepSeek-MoE-16B
2026.03
51.05
Feedback
Search any
task
Search any
task