# Security Tensors as a Cross-Modal Bridge: Extending Text-Aligned Safety to Vision in LVLM

arXiv: https://arxiv.org/abs/2507.20994 · HTML: https://mlsanity.com/paper/2507.20994/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 17 of its 27 references that are in the map.
- First posted on arXiv 28 Jul 2025.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM jailbreaks and adversarial robustness](https://mlsanity.com/mllm/mllm-jailbreaks-and-adversarial-robustness/)
- LLM alignment / RLHF / preference optimisation / [Safety alignment and jailbreak attacks](https://mlsanity.com/alignment/safety-alignment-and-jailbreak-attacks/)

## Builds on (most cited first, 17 of 17)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [Teaching Small Language Models to Reason](https://mlsanity.com/paper/2212.08410/) (2022)
- [Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models](https://mlsanity.com/paper/2402.02207/) (2024)
- [On the Adversarial Robustness of Multi-Modal Foundation Models](https://mlsanity.com/paper/2308.10741/) (2023)
- [Safety Layers in Aligned Large Language Models: The Key to LLM Security](https://mlsanity.com/paper/2408.17003/) (2024)
- [Jailbreak Vision Language Models via Bi-Modal Adversarial Prompt](https://mlsanity.com/paper/2406.04031/) (2024)
- [DRESS: Instructing Large Vision-Language Models to Align and Interact with Humans via Natural Language Feedback](https://mlsanity.com/paper/2311.10081/) (2023)
- [Enhancing Visual-Language Modality Alignment in Large Vision Language Models via Self-Improvement](https://mlsanity.com/paper/2405.15973/) (2024)
- [Break the Visual Perception: Adversarial Attacks Targeting Encoded Visual Tokens of Large Vision-Language Models](https://mlsanity.com/paper/2410.06699/) (2024)
- [Cross-Modal Safety Mechanism Transfer in Large Vision-Language Models](https://mlsanity.com/paper/2410.12662/) (2024)
- [How Does Vision-Language Adaptation Impact the Safety of Vision Language Models?](https://mlsanity.com/paper/2410.07571/) (2024)
- [Spot Risks Before Speaking! Unraveling Safety Attention Heads in Large Vision-Language Models](https://mlsanity.com/paper/2501.02029/) (2025)
- [Tracking the Copyright of Large Vision-Language Models through Parameter Learning Adversarial Images](https://mlsanity.com/paper/2502.16593/) (2025)

## Abstract (arXiv)

Large visual-language models (LVLMs) integrate aligned large language models (LLMs) with visual modules to process multimodal inputs. However, the safety mechanisms developed for text-based LLMs do not naturally extend to visual modalities, leaving LVLMs vulnerable to harmful image inputs. To address this cross-modal safety gap, we introduce security tensors - trainable input vectors applied during inference through either the textual or visual modality. These tensors transfer textual safety alignment to visual processing without modifying the model's parameters. They are optimized using a curated dataset containing (i) malicious image-text pairs requiring rejection, (ii) contrastive benign pairs with text structurally similar to malicious queries, with the purpose of being contrastive examples to guide visual reliance, and (iii) general benign samples preserving model functionality. Experimental results demonstrate that both textual and visual security tensors significantly enhance LVLMs' ability to reject diverse harmful visual inputs while maintaining near-identical performance on benign tasks. Further internal analysis towards hidden-layer representations reveals that security tensors successfully activate the language module's textual "safety layers" in visual inputs, thereby effectively extending text-based safety to the visual modality.