DOE OSTI · code-154189
Double Visual Defense
Abstract
This is the official code for the paper "Double Visual Defense: Adversarial Pre-training and Instruction Tuning for Improving Vision-Language Model Robustness". This code can be used to produce vision language models (VLMs), like LLaVA, with enhanced robustness to adversarial attacks (e.g. jailbreaks).
Keep this discovery
Explore connections, maps & timelines
Bartoldson, Brian [Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)], Wang, Zeyu [Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)], Kailkhura, Bhavya [Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)]. 2024-12-12. Double Visual Defense. https://doi.org/10.11578/dc.20250416.3
Cite the original work for its findings. Save a collection to share your selection of sources.