Diagnostics for Deep Neural Networks with Automated Copy/Paste Attacks

2022-11-18 · Stephen Casper, Kaivalya Hariharan, Dylan Hadfield-Menell

This paper considers the problem of helping humans exercise scalable oversight over deep neural networks (DNNs). Adversarial examples can be useful by helping to reveal weaknesses in DNNs, but they can be difficult to interpret or draw actionable conclusions from. Some previous works have proposed using human-interpretable adversarial attacks including copy/paste attacks in which one natural image pasted into another causes an unexpected misclassification. We build on these with two contributions. First, we introduce Search for Natural Adversarial Features Using Embeddings (SNAFUE) which offers a fully automated method for finding copy/paste attacks. Second, we use SNAFUE to red team an ImageNet classifier. We reproduce copy/paste attacks from previous works and find hundreds of other easily-describable vulnerabilities, all without a human in the loop. Code is available at https://github.com/thestephencasper/snafue

📄 PDF Abstract BibTeX arXiv:2211.10024

Code (1)

thestephencasper/snafue 공식 구현 pytorch

Copy-Paste to Mitigate Large Language Model Hallucinations

2025-10-01 · Yongchao Long, Xian Wu, Yingying Zhang, Xianbin Wen 외 arxiv

While Retrieval-Augmented Generation (RAG) enables large language models (LLMs) to generate contextually grounded responses, contextual faithfulness remains challenging as LLMs may not consistently trust provided context…

Enrich the content of the image Using Context-Aware Copy Paste

2024-07-11 · Qiushi Guo

Data augmentation remains a widely utilized technique in deep learning, particularly in tasks such as image classification, semantic segmentation, and object detection. Among them, Copy-Paste is a simple yet effective me…

Data Augmentation Diversity image-classification Image Classification+3

CopyPaste: An Augmentation Method for Speech Emotion Recognition

2020-10-27 · Raghavendra Pappagari, Jesús Villalba, Piotr Żelasko, Laureano Moro-Velazquez 외

Data augmentation is a widely used strategy for training robust machine learning models. It partially alleviates the problem of limited data for tasks like speech emotion recognition (SER), where collecting data is expen…

Data Augmentation Emotion Recognition Speaker Recognition Speech Emotion Recognition+1

X-Paste: Revisiting Scalable Copy-Paste for Instance Segmentation using CLIP and StableDiffusion

2022-12-07 · Hanqing Zhao, Dianmo Sheng, Jianmin Bao, Dongdong Chen 외

Copy-Paste is a simple and effective data augmentation strategy for instance segmentation. By randomly pasting object instances onto new background images, it creates new training data for free and significantly boosts t…

Data Augmentation Instance Segmentation Object Object Detection+4

Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation

2020-12-13 · CVPR 2021 1 · Golnaz Ghiasi, Yin Cui, Aravind Srinivas, Rui Qian 외

Building instance segmentation models that are data-efficient and can handle rare object categories is an important challenge in computer vision. Leveraging data augmentations is a promising direction towards addressing …

Data Augmentation Image Augmentation Instance Segmentation Object Detection+2