JailbreakBench

JailbreakBench: An Open Robustness Benchmark for Jailbreaking Language Models Paper: JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models Data: JailbreaBench-HFLink About Jailbreakbench is an open-source robustness benchmark for jailbreaking large language models (LLMs). The goal of this

Papers using JailbreakBench (4)

Reflect-Guard: Enhancing LLM Safeguards against Adversarial Prompts via Logical Self-Reflection2026Analysing the Safety Pitfalls of Steering Vectors2026BreakFun: Jailbreaking LLMs via Schema Exploitation2025LARGO: Latent Adversarial Reflection Through Gradient Optimization For Jailbreaking Llms2025