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

Response Classification on BeaverTails V Text-Image Response

84.8F1 Score

Qwen3Guard-8B-Gen

55.88863.39470.978.406Dec 29, 2025Jan 5, 2026Jan 12, 2026Jan 19, 2026Jan 26, 2026Feb 2, 2026Feb 10, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
84.8
2026.02
84.6
2026.02
84.5
2026.02
84.2
2026.02
84.1
84.02
2026.02
84
2026.02
83.4
83.1
2026.02
82.9
2026.02
82.7
82.35
81.73
2026.02
81.2
81.13
80.72
2026.02
80.6
80.44
79.83
79.19
77.85
2026.02
77.4
74.62
73.51
73.39
2025.12
72.92
72.59
72.4
70.91
69.49
2026.02
67.5
2026.02
67.3
2026.02
66.8
66.62
66.13
2025.12
63.5
62.8
2026.02
61.1
57