Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multi-class Classification on MMLU HELM

59.48Balanced Accuracy

Base-Best

34.551241.023147.49553.9669Feb 14, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
59.48
2026.02
58.8
2026.02
57.73
2026.02
55.1
2026.02
54.99
2026.02
54.98
2026.02
54.97
2026.02
54.97
2026.02
54.97
2026.02
54.9
2026.02
54.82
2026.02
54.74
2026.02
54.6
2026.02
54.58
2026.02
54.42
53.9
2026.02
35.51