OOD safety category inference (Stage 2)
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
26.86Mean Reward
4
Feb 27, 2026
25.95Mean Reward
4
Feb 27, 2026
13.28Reward Mean
4
Feb 27, 2026
32.94Reward-mean
4
Feb 27, 2026
36.45Mean Reward
4
Feb 27, 2026
23.08Reward Mean
4
Feb 27, 2026
—
—Reward Mean
0
Feb 26, 2026