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

Knowledge-Based QA on Global-MMLU-Lite (8 Seen Languages)

65.2MMLU Score

CAGrad

36.0843.6451.258.76Jun 4, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
65.2
2026.06
64.8
2026.06
64.6
2026.06
63.4
2026.06
61.7
2026.06
58.4
2026.06
58.1
2026.06
58
2026.06
56.9
2026.06
55.9
2026.06
45.6
2026.06
45.5
2026.06
45.2
2026.06
45.2
2026.06
44.8
2026.06
44.2
2026.06
43.9
2026.06
42.5
2026.06
41.8
2026.06
37.2