LMVD-ID: 5eceb158
Published November 1, 2024

RL-Tuned LLM Jailbreak

model-layer
jailbreak
blackbox
fine-tuning
safety
Affected Models:llama2-7b-chat, claude 2, gpt-3.5, gemma-2b-it, vicuna-7b, gpt-4

Research Paper

LLMStinger: Jailbreaking LLMs using RL fine-tuned LLMs

View Paper

© 2025 Promptfoo. All rights reserved.