paper-with-me

홈 › Papers

NaturalAdversaries: Can Naturalistic Adversaries Be as Effective as Artificial Adversaries?

2022-11-08 · Saadia Gabriel, Hamid Palangi, Yejin Choi

While a substantial body of prior work has explored adversarial example generation for natural language understanding tasks, these examples are often unrealistic and diverge from the real-world data distributions. In this work, we introduce a two-stage adversarial example generation framework (NaturalAdversaries), for designing adversaries that are effective at fooling a given classifier and demonstrate natural-looking failure cases that could plausibly occur during in-the-wild deployment of the models. At the first stage a token attribution method is used to summarize a given classifier's behaviour as a function of the key tokens in the input. In the second stage a generative model is conditioned on the key tokens from the first stage. NaturalAdversaries is adaptable to both black-box and white-box adversarial attacks based on the level of access to the model parameters. Our results indicate these adversaries generalize across domains, and offer insights for future research on improving robustness of neural text classification models.

📄 PDF Abstract BibTeX arXiv:2211.04364

Code (0)

등록된 구현이 없습니다.

Tasks

Natural Language Understandingtext-classificationText Classification

Similar Papers 제목 키워드 기반

Enhancing Resilience of Deep Learning Networks by Means of Transferable Adversaries

2020-05-27 · Moritz Seiler, Heike Trautmann, Pascal Kerschke

Artificial neural networks in general and deep learning networks in particular established themselves as popular and powerful machine learning algorithms. While the often tremendous sizes of these networks are beneficial…

AI-Driven Adaptive Adversaries and the Erosion of Cryptographic Trust in Public Key Systems

2026-05-23 · Petar Radanliev arxiv

This paper examines the erosion of Public Key Cryptography (PKC) security under adaptive adversarial optimisation driven by artificial intelligence. The problem addressed is the growing mismatch between algorithm-centric…

On the power of adaptivity in statistical adversaries

2021-11-19 · Guy Blanc, Jane Lange, Ali Malik, Li-Yang Tan

We study a fundamental question concerning adversarial noise models in statistical problems where the algorithm receives i.i.d. draws from a distribution $\mathcal{D}$. The definitions of these adversaries specify the ty…

All

Combining Adversaries with Anti-adversaries in Training

2023-04-25 · Xiaoling Zhou, Nan Yang, Ou wu

Adversarial training is an effective learning technique to improve the robustness of deep neural networks. In this study, the influence of adversarial training on deep learning models in terms of fairness, robustness, an…

FairnessMeta-Learning

Improving Local Effectiveness for Global Robustness Training

2021-01-01 · Jingyue Lu, M. Pawan Kumar

Despite its increasing popularity, deep neural networks are easily fooled. Toalleviate this deficiency, researchers are actively developing new training strategies,which encourage models that are robust to small input…