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

High-stakes specialized classification on Claude9 (test)

50.31Macro F1

LLM-Annotation

5.943617.461828.9840.4982Jun 1, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.06
50.31
2026.06
48.9
2026.06
46.6
2026.06
45
2026.06
42.59
2026.06
39.08
2026.06
38.27
2026.06
36.45
2026.06
36.39
2026.06
35.79
2026.06
35.28
2026.06
34.69
2026.06
34.13
2026.06
33.2
2026.06
33.2
2026.06
33.2
2026.06
33.2
2026.06
32.4
2026.06
25.26
2026.06
22.07
2026.06
21.6
2026.06
21.11
2026.06
19.41
2026.06
19.09
2026.06
18.67
2026.06
16.33
2026.06
16.1
2026.06
15
2026.06
14.58
2026.06
12.03
2026.06
11.31
2026.06
10.57
2026.06
10.57
2026.06
10.57
2026.06
10.57
2026.06
10.53
2026.06
7.65