Overrefusal evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
99.85RR
32
Jun 15, 2026
97.6Helpfulness Score
12
Apr 13, 2026
9.9Overrefusal Rate (OB)
7
Apr 23, 2026
0.3Refusal Rate (RR)
6
Mar 13, 2026
0RR
6
Mar 13, 2026
92RR
6
Mar 13, 2026
4.44Refusal Rate
6
Mar 13, 2026
91Performance Score
2
Mar 12, 2026
1Performance Score
2
Mar 12, 2026