paper-with-me

Papers

Self-interpreting Adversarial Images

2024-07-12 · Tingwei Zhang, Collin Zhang, John X. Morris, Eugene Bagdasarian, Vitaly Shmatikov

We introduce a new type of indirect, cross-modal injection attacks against visual language models that enable creation of self-interpreting images. These images contain hidden "meta-instructions" that control how models answer users' questions about the image and steer models' outputs to express an adversary-chosen style, sentiment, or point of view. Self-interpreting images act as soft prompts, conditioning the model to satisfy the adversary's (meta-)objective while still producing answers based on the image's visual content. Meta-instructions are thus a stronger form of prompt injection. Adversarial images look natural and the model's answers are coherent and plausible, yet they also follow the adversary-chosen interpretation, e.g., political spin, or even objectives that are not achievable with explicit text instructions. We evaluate the efficacy of self-interpreting images for a variety of models, interpretations, and user prompts. We describe how these attacks could cause harm by enabling creation of self-interpreting content that carries spam, misinformation, or spin. Finally, we discuss defenses.

📄 PDF Abstract BibTeX arXiv:2407.08970

Code (1)

tingwei-zhang/soft-prompts-go-hard 공식 구현 pytorch

Tasks

Misinformation

Similar Papers 제목 키워드 기반

Boosting Generative Adversarial Transferability with Self-supervised Vision Transformer Features

2025-06-26 · Shangbo Wu, Yu-an Tan, Ruinan Ma, Wencong Ma 외

The ability of deep neural networks (DNNs) come from extracting and interpreting features from the data provided. By exploiting intermediate features in DNNs instead of relying on hard labels, we craft adversarial pertur…

Contrastive LearningSelf-Supervised Learning

On Visual Hallmarks of Robustness to Adversarial Malware

2018-05-09 · Alex Huang, Abdullah Al-Dujaili, Erik Hemberg, Una-May O'Reilly

A central challenge of adversarial learning is to interpret the resulting hardened model. In this contribution, we ask how robust generalization can be visually discerned and whether a concise view of the interactions be…

Hamming Similarity and Graph Laplacians for Class Partitioning and Adversarial Image Detection

2023-05-02 · Huma Jamil, Yajing Liu, Turgay Caglar, Christina M. Cole 외

Researchers typically investigate neural network representations by examining activation outputs for one or more layers of a network. Here, we investigate the potential for ReLU activation patterns (encoded as bit vector…

Towards Interpreting Recurrent Neural Networks through Probabilistic Abstraction

2019-09-22 · Guoliang Dong, Jingyi Wang, Jun Sun, Yang Zhang 외

Neural networks are becoming a popular tool for solving many real-world problems such as object recognition and machine translation, thanks to its exceptional performance as an end-to-end solution. However, neural networ…

Machine TranslationObject Recognition

Interpreting Adversarial Examples with Attributes

2019-04-17 · Sadaf Gulshad, Jan Hendrik Metzen, Arnold Smeulders, Zeynep Akata

Deep computer vision systems being vulnerable to imperceptible and carefully crafted noise have raised questions regarding the robustness of their decisions. We take a step back and approach this problem from an orthogon…

AttributeGeneral Classification