AdvBench

AdvBench is a benchmark used to evaluate the safety and trustworthiness of large language models by assessing their responses to potentially harmful content.

Papers using AdvBench (2)

Federated Fine-Tuning of Large Language Models: Kahneman-Tversky vs. Direct Preference Optimization2025 · 2 citesResponsible Federated LLMs via Safety Filtering and Constitutional AI2025