Cross-Modal Safety Mechanism Transfer in Large Vision-Language Models
This addresses safety vulnerabilities in vision-language models for users relying on multimodal AI systems, though it is incremental as it builds on existing alignment methods.
The paper tackled the problem that existing vision-language alignment methods fail to transfer text safety mechanisms to vision in Large Vision-Language Models, leading to vulnerabilities with toxic images, and proposed a Text-Guided Alignment method that successfully transfers safety mechanisms without fine-tuning while maintaining general performance.
Vision-language alignment in Large Vision-Language Models (LVLMs) successfully enables LLMs to understand visual input. However, we find that existing vision-language alignment methods fail to transfer the existing safety mechanism for text in LLMs to vision, which leads to vulnerabilities in toxic image. To explore the cause of this problem, we give the insightful explanation of where and how the safety mechanism of LVLMs operates and conduct comparative analysis between text and vision. We find that the hidden states at the specific transformer layers play a crucial role in the successful activation of safety mechanism, while the vision-language alignment at hidden states level in current methods is insufficient. This results in a semantic shift for input images compared to text in hidden states, therefore misleads the safety mechanism. To address this, we propose a novel Text-Guided vision-language Alignment method (TGA) for LVLMs. TGA retrieves the texts related to input vision and uses them to guide the projection of vision into the hidden states space in LLMs. Experiments show that TGA not only successfully transfers the safety mechanism for text in basic LLMs to vision in vision-language alignment for LVLMs without any safety fine-tuning on the visual modality but also maintains the general performance on various vision tasks (Safe and Good).