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

Granularity Estimation on GRANOLA-EQ (test)

69.84Kendall's τ

GPT-4.1 mini

1.522419.258736.99554.7313May 26, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
69.8474.0488.24
2026.05
55.5369.6978.06
2026.05
54.6668.9177.66
2026.05
54.2867.5876.3
2026.05
54.0467.5574.05
2026.05
53.1666.0476.7
2026.05
52.5166.6773.65
2026.05
52.2865.1372.93
2026.05
50.752.3775.73
2026.05
50.5364.4169.49
2026.05
28.7937.8741.14
2026.05
28.7337.7642.31
2026.05
28.6536.9439
2026.05
25.5428.1561.15
2026.05
4.151.077.91