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

Sabotage Detection on TRACE

90.1Average Precision

Monitor

52.45262.2267281.774Apr 13, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.04
90.1
2026.04
88.2
2026.04
88.1
2026.04
87.4
2026.04
86.1
2026.04
84.9
2026.04
84.3
2026.04
83
2026.04
82.2
2026.04
81.9
2026.04
80.8
2026.04
77.2
2026.04
75.4
2026.04
75.1
2026.04
74.3
2026.04
71.7
2026.04
71.4
2026.04
70
2026.04
70
2026.04
68
2026.04
66.1
2026.04
63.8
2026.04
63.4
2026.04
62.1
2026.04
59.3
2026.04
56.4
2026.04
54.6
2026.04
53.9