Long-term state poisoning evaluation
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
3.66Harm Score (HS)
8
May 11, 2026
3.82Harm Score (HS)
4
May 11, 2026
3.76Harm Score (HS)
4
May 11, 2026
4.35Harm Score (HS)
4
May 11, 2026
4.03Harm Score (HS)
4
May 11, 2026
4.34Harm Score (HS)
4
May 11, 2026
3.9Harm Score (HS)
4
May 11, 2026