Edoardo Debenedetti

A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents.

RobustBench: a standardized adversarial robustness benchmark [NeurIPS 2021 Benchmarks and Datasets Track]

JailbreakBench: An Open Robustness Benchmark for Jailbreaking Language Models [NeurIPS 2024 Datasets and Benchmarks Track]