Task-driven attention attack: An enhanced adversarial framework for long text

Q. N. Xu

doi:10.1016/j.eswa.2025.128834

2025-06-29

PRE

Q. N. Xu

Jianxin Bi

Jingyuan Liu

DOI：10.1016/j.eswa.2025.128834

原文链接

原文求助

摘要

En 中文

• 现有攻击因评分机制效率低下而在长文本上失败。 • TDAA：面向长文本的新型任务驱动攻击框架。 • 整合了上下文语义与任务特定注意力。 • 自适应长度阈值 + 统计上非必要句子的消除。 • 相比基线，ASR提升34.7%，效率提升40%。

期刊

Expert Systems with Applications

IF:

7.5

论文数:

3.0W

被引数:

10.2W

机构

暂无机构信息

引用论文

Contextualized Perturbation for Textual Adversarial Attack

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

IF0

2021-01-01

OAAI

Dianqi Li; Yizhe Zhang; Hao Peng; Liqun Chen; Chris Brockett; Ming-Ting Sun; Bill Dolan

Long Short-Term Memory 长短期记忆

Neural Computation

IF0

1997-11-01

PREAI

Sepp Hochreiter; Jürgen Schmidhuber

Self-Attention Attribution: Interpreting Information Interactions Inside Transformer

Proceedings of the AAAI Conference on Artificial Intelligence

IF0

2021-05-18

OAAI

Yaru Hao; Li Dong; Furu Wei; Ke Xu

Aliasing black box adversarial attack with joint self-attention distribution and confidence probability

EXPERT SYSTEMS WITH APPLICATIONS

IF7.5

2023-03-01

PREAI

Liu, Jun; Jin, Haoyu; Xu, Guangxia; Lin, Mingwei; Wu, Tao; Nour, Majid; Alenezi, Fayadh; Alhudhaif, Adi; Polat, Kemal

WordChange: Adversarial Examples Generation Approach for Chinese Text Classification

IEEE ACCESS

IF3.6

2020-01-01

OAAI

Nuo, Cheng; Chang, Guo-Qin; Gao, Haichang; Pei, Ge; Zhang, Yang

TextJuggler: Fooling text classification tasks by generating high-quality adversarial examples TextJuggler: 通过生成高质量的对抗示例来愚弄文本分类任务

KNOWLEDGE-BASED SYSTEMS

IF7.6

2024-09-01

PREAI

Peng, Hao; Wang, Zhe; Wei, Chao; Zhao, Dandan; Xu, Guangquan; Han, Jianming; Guo, Shixin; Zhong, Ming; Ji, Shouling

没有更多内容