arXiv Machine Learning
Seeing Through Conflicts: Improving Instruction Hierarchy Alignment in Vision-Language Models
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Sep 7
MM-IFEval-Pro: A Multilingual and Attack-Resistant Benchmark for Instruction-Following in Vision-Language Models
MM-IFEval-Pro is a new multilingual benchmark for evaluating instruction-following in vision-language models, covering both Chinese and English tasks. It includes 4 major task categories, 24 subcategories, and 8 instruction categories with 52 subcategories, each sample featuring an average of 3.0 constraints to mimic complex instruction scenarios. A reinforcement-learning training set with Chinese and adversarial instructions improves model performance on MM-IFEval-Pro and transfers well to other multimodal benchmarks, showing strong cross-task and cross-language generalization.
arXiv AI
Aug 12
SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
arXiv:2608. 10513v1 Announce Type: cross Abstract: Large vision-language models (LVLMs) remain vulnerable to jailbreak attacks that exploit visual inputs to bypass safety alignment inherited from their language backbones.
arXiv AI
Jul 22
Dual Adversarial Fine-tuning for Enhancing Robustness of Large Vision Language Model
arXiv:2607. 18958v1 Announce Type: cross Abstract: While Large Vision-Language Models (LVLMs), represented by LLaVA and GPT-4V, have demonstrated remarkable capabilities, their visual inputs remain vulnerable to adversarial attacks, posing significant security risks.
Hugging Face Trending Papers
Aug 19
Breaking the weakest link to evade vision language models
The paper investigates how Vision Language Models (VLMs) can be fooled by tiny, human‑imperceptible changes to images. It introduces a gradient‑based attack that targets only the vision encoder, reducing computational cost while still effectively disrupting both untargeted and targeted multimodal interpretations. Experiments on open‑source VLMs such as Qwen2.5‑VL, Granite‑Vision, FastVLM, and Phi‑3.5‑Vision demonstrate that these small perturbations can dramatically alter the models’ textual outputs.
Hugging Face Trending Papers
Aug 11
SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
Large vision-language models (LVLMs) remain vulnerable to jailbreak attacks that exploit visual inputs to bypass safety alignment inherited from their language backbones. We propose SafeCap, a reinforcement-learning framework that aligns LVLMs through learned self-captioning.
Hugging Face Trending Papers
Jun 2
Investigating Adversarial Robustness of Multi-modal Large Language Models
Multi-modal Large Language Models (MLLMs) achieve strong performance on vision-language tasks, but incorporating visual inputs through a vision encoder (e. g.