Safety and Informativeness Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
85.3Drugs & Weapons Safety Score
6
Jun 2, 2026
75Deception Rate (Safe)
4
Jun 2, 2026
90Safety Score
4
Jun 2, 2026
95Safety Score
4
Jun 2, 2026
80Safety Score
4
Jun 2, 2026
76.7Safety Score
4
Jun 2, 2026
100Safety Score
4
Jun 2, 2026
90Safety Score
4
Jun 2, 2026
85Safety Score
4
Jun 2, 2026
95Safety Rate
4
Jun 2, 2026
100Safety Score
4
Jun 2, 2026
91Safety Rate
4
Jun 2, 2026