paper-with-me

홈 › Papers

Learning To See But Forgetting To Follow: Visual Instruction Tuning Makes LLMs More Prone To Jailbreak Attacks

2024-05-07 · Georgios Pantazopoulos, Amit Parekh, Malvina Nikandrou, Alessandro Suglia

Augmenting Large Language Models (LLMs) with image-understanding capabilities has resulted in a boom of high-performing Vision-Language models (VLMs). While studying the alignment of LLMs to human values has received widespread attention, the safety of VLMs has not received the same attention. In this paper, we explore the impact of jailbreaking on three state-of-the-art VLMs, each using a distinct modeling approach. By comparing each VLM to their respective LLM backbone, we find that each VLM is more susceptible to jailbreaking. We consider this as an undesirable outcome from visual instruction-tuning, which imposes a forgetting effect on an LLM's safety guardrails. Therefore, we provide recommendations for future work based on evaluation strategies that aim to highlight the weaknesses of a VLM, as well as take safety measures into account during visual instruction tuning.

📄 PDF Abstract BibTeX arXiv:2405.04403

Code (1)

gpantaz/vl_jailbreak 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Separable Mixture of Low-Rank Adaptation for Continual Visual Instruction Tuning

2024-11-21 · Ziqi Wang, Chang Che, Qi Wang, Yangyang Li 외

Visual instruction tuning (VIT) enables multimodal large language models (MLLMs) to effectively handle a wide range of vision tasks by framing them as language-based instructions. Building on this, continual visual instr…

Continual LearningInstruction Following

Actions as Language: Fine-Tuning VLMs into VLAs Without Catastrophic Forgetting

2025-09-26 · Asher J. Hancock, Xindi Wu, Lihan Zha, Olga Russakovsky 외 arxiv

Fine-tuning vision-language models (VLMs) on robot teleoperation data to create vision-language-action (VLA) models is a promising paradigm for training generalist policies, but it suffers from a fundamental tradeoff: le…

Visual Question AnsweringZero-shot GeneralizationInstruction Following

Refine Large Language Model Fine-tuning via Instruction Vector

2024-06-18 · Gangwei Jiang, Zhaoyi Li, Defu Lian, Ying WEI

Fine-tuning large language models (LLMs) can cause them to lose their general capabilities. However, the intrinsic mechanisms behind such forgetting remain unexplored. In this paper, we begin by examining this phenomenon…

Instruction FollowingLanguage ModelingLanguage ModellingLarge Language Model

Continual Instruction Tuning for Large Multimodal Models

2023-11-27 · Jinghan He, Haiyun Guo, Ming Tang, Jinqiao Wang

Instruction tuning is now a widely adopted approach to aligning large multimodal models (LMMs) to follow human intent. It unifies the data format of vision-language tasks, enabling multi-task joint training. However, vis…

Continual Learning

Layer-Aware Task Arithmetic: Disentangling Task-Specific and Instruction-Following Knowledge

2025-02-27 · Yan-Lun Chen, Yi-Ru Wei, Chia-Yi Hsu, Chia-Mu Yu 외

Large language models (LLMs) demonstrate strong task-specific capabilities through fine-tuning, but merging multiple fine-tuned models often leads to degraded performance due to overlapping instruction-following componen…

GSM8KHumanEvalInstruction FollowingMulti-Task Learning+1