Question Answering on MMLU Computer Science Large Shift (test) benchmark leaderboard · SOTA2 Research

Question Answering on MMLU Computer Science Large Shift (test)

46.3AccuracyBayesian-LoRA

Evaluation Results

Method	Links
Bayesian-LoRA2026.01		46.3	11.1	1.22
LLLA2026.01		45.6	30.3	1.28
LA2026.01		45.1	14.5	1.26
BLoB2026.01		44.6	12.6	1.38
Temp2026.01		43.5	16.4	1.31
MAP2026.01		42	34.2	1.9
Dropout2026.01		41.9	33.7	1.86
Ckpt Ens2026.01		41.1	29.1	1.55
BBB2026.01		40.5	18.6	1.49