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

Multi-turn MLLM Safety Evaluation on STEER-SPA Multi-turn

66.67Safety

SAFER-STEER

3.635620.000336.36552.7297Mar 18, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
66.6770.93
2026.03
61.1361.13
2026.03
54.7666.27
2026.03
51.3265.66
2026.03
48.6858.11
2026.03
41.1352.08
2026.03
39.8552.87
2026.03
32.5822.35
2026.03
29.0146.56
2026.03
28.2422.14
2026.03
25.8633.84
2026.03
2550.42
2026.03
17.1135.36
2026.03
16.2918.18
2026.03
14.8923.28
2026.03
10.5721.13
2026.03
8.7123.48
2026.03
6.0610.98