AdvBench

Dataset Card for AdvBench Paper: Universal and Transferable Adversarial Attacks on Aligned Language Models Data: AdvBench Dataset About AdvBench is a set of 500 harmful behaviors formulated as instructions. These behaviors range over the same themes as the harmful strings setting, but the adversary’s goal is instead to

Papers using AdvBench (4)

Safety Targeted Embedding Exploit via Refinement2026The Resurgence of GCG Adversarial Attacks on Large Language Models2025Intrinsic Model Weaknesses: How Priming Attacks Unveil Vulnerabilities in Large Language Models2025Responsible Federated LLMs via Safety Filtering and Constitutional AI2025