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

Common Sense Reasoning on Six Benchmarks Suite (ARC-e, PIQA, OpenbookQA, Winogrande, HellaSwag, MathQA)

61Average Accuracy

Original

30.8438.6746.554.33Apr 2, 2026
Updated 16d ago

Evaluation Results

MethodLinks
2026.04
61
2026.04
57
2026.04
56
2026.04
56
2026.04
55
2026.04
54
2026.04
53
2026.04
52
2026.04
51
2026.04
50
2026.04
42
2026.04
41
2026.04
36
2026.04
32
2026.04
32