相关实验视频

Updated: Jul 9, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

适应性WordBug:生成与深度学习分类器相对应的适应性评分策略的对抗性文本

Yunting Zhang¹, Lin Ye¹, Baisong Li²

¹School of Cyberspace Science, Harbin Institute of Technology, Harbin, 150001, China.

Neural networks : the official journal of the International Neural Network Society

|November 8, 2025

概括

通过结合多种单词重要性评分方法,AdaptiveWordBug增强了对抗性文本生成. 这种适应性方法提高了针对BERT和ChatGPT等深度学习模型的攻击效率.

科学领域:

自然语言处理自然语言处理.
机器学习安全机器学习安全

背景情况:

深度学习模型容易受到敌对攻击.
目前的单词级攻击方法依赖于单个模型依赖的评分,限制了准确性和全面性.
有效的对抗性文本生成对于开发强大的防御机制至关重要.

研究的目的:

为文本分类提出一种新的黑子对抗性文本生成方法.
在对抗性攻击中提高单词重要性评分的准确性和全面性.
开发一种可适应各种目标模型的多功能方法.

主要方法:

推出了AdaptiveWordBug,一种黑子对抗性文本生成方法.
开发了自适应评分策略 (ASS),结合了四种评分方法 (三种依赖模型,一种独立于模型).
对每个评分方法实施的自适应参数,可根据文本进行调整.

主要成果:

与基线方法相比,AdaptiveWordBug显示出更高的攻击效率.
适应性评分策略准确地识别了不同文本中的重要单词.
该方法对包括中国BERT和ChatGPT在内的多种目标模型具有很好的适用性.

结论:

通过其自适应评分策略,AdaptiveWordBug显著提高了对抗性文本生成效率.
拟议的方法提供了一种更全面,更准确的方法来识别干扰的重要词.
适应WordBug提供了一个灵活和有效的工具,用于评估和提高文本分类模型的稳定性.

关键词:

对抗性的例子.深度神经网络是一个神经网络.评分方法评分方法文字分类文本分类文本分类文字对抗性攻击是文字对抗性的攻击.

相关概念视频

Non-equilibrium in the Cell

Non-equilibrium in the Cell

An important concept in studying metabolism and energy is that of chemical equilibrium. Most chemical reactions are reversible. They can proceed in both directions, releasing energy into their environment in one direction, and absorbing it from the environment in the other direction. The same is true for the chemical reactions involved in cell metabolism, such as the breaking down and building up of proteins into and from individual amino acids, respectively. Reactants within a closed system...

Force Classification

Force Classification

Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...