List your company

Abstention Trigger on Tool selection benchmark

0.88Certified Upper Bound (pu)Anomaly mon.
0.7240.76450.8050.8455Oct 5, 2025

Evaluation results

MethodLinks
Anomaly mon.Mitigation=Anomaly mon., Selector model=Llama-3.1, Generator model=Gemma-32025.100.88
SecAlignMitigation=SecAlign, Selector model=Llama-3.1, Generator model=Gemma-32025.100.84
StruQMitigation=StruQ, Selector model=Llama-3.1, Generator model=Gemma-32025.100.79
PerplexityMitigation=Perplexity, Selector model=Llama-3.1, Generator model=Gemma-32025.100.75
Defended retr.Mitigation=Defended retr., Selector model=Llama-3.1, Generator model=Gemma-32025.100.74
NoneMitigation=None, Selector model=Llama-3.1, Generator model=Gemma-32025.100.73