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

Multiple Choice Question Answering on MMLU (Performance Change Tracking)

58MMLU Accuracy Gain (Absolute)

FLOATDOOR

-1.363214.048429.4644.8716Feb 11, 2026Mar 4, 2026Mar 25, 2026Apr 15, 2026May 6, 2026May 27, 2026Jun 17, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
58--
2026.06
35--
2026.06
28--
2026.06
7--
2026.06
2--
2026.02
0.92-62.33
2026.02
-61.41-