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

Evaluation Benchmarks Zero-shot

71.76Average Accuracy

Dense

40.164848.367456.5764.7726May 23, 2025Jul 23, 2025Sep 22, 2025Nov 23, 2025Jan 23, 2026Mar 25, 2026May 26, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2025.05
71.76
2026.05
70.9
2026.05
70.2
2026.05
69.08
2026.05
67.51
2026.05
66.1
2026.05
65.74
2025.05
65.11
2025.05
61.79
2025.05
60.84
2026.05
60.1
2026.05
59.33
2026.05
57.96
2026.05
57.67
2025.05
57.57
2025.05
56.77
2026.05
54.91
2026.05
54.7
2026.05
54.64
2026.05
54.02
2026.05
53.97
2026.05
53.92
2026.05
53.44
2025.05
52.17
2025.05
50.83
2025.05
48.05
2025.05
47.59
2026.05
46.29
2026.05
46.13
2026.05
44.8
2026.05
43.28
2026.05
42.31
2026.05
42
2026.05
41.38