arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2109.06777cs.LGcs.SI
收藏

PETGEN:基于深度序列嵌入分类模型的个性化文本生成攻击

PETGEN: Personalized Text Generation Attack on Deep Sequence Embedding-based Classification Models

Georgia Institute of Technology(佐治亚理工学院)

Bing He, Mustaque Ahamad, Srijan Kumar

摘要:

恶意用户接下来应该写什么来欺骗检测模型?识别恶意用户对于确保互联网平台的安全性和完整性至关重要。已经创建了几个基于深度学习的检测模型。然而,恶意用户可以通过操纵他们的行为来逃避深度检测模型,使这些模型几乎没有用处。这种深度检测模型对对抗攻击的脆弱性是未知的。在这里,我们创建了一个新的对抗攻击模型,针对基于深度用户序列嵌入的分类模型,使用用户帖子的序列来生成用户嵌入并检测恶意用户。在攻击中,对手生成新的帖子来欺骗分类器。我们提出了一种新的端到端个性化文本生成攻击模型,称为PEGEN,同时降低了检测模型的效率,并生成具有几个关键理想属性的帖子。具体而言,PETGEN生成的帖子是个性化的用户的写作风格,具有关于给定的目标上下文的知识,知道用户的历史帖子的目标上下文,并封装用户的最近的主题兴趣。我们在两个真实世界的数据集(Yelp和Wikipedia,都有恶意用户的真实数据)上进行了广泛的实验,以表明PEGEN显著降低了流行的基于深度用户序列嵌入的分类模型的性能。PEGEN在白盒和黑盒分类器设置中的文本质量和攻击效率方面优于五个攻击基线。总的来说,这项工作铺平了道路,下一代的对手意识的序列分类模型。

英文摘要:

What should a malicious user write next to fool a detection model? Identifying malicious users is critical to ensure the safety and integrity of internet platforms. Several deep learning-based detection models have been created. However, malicious users can evade deep detection models by manipulating their behavior, rendering these models of little use. The vulnerability of such deep detection models against adversarial attacks is unknown. Here we create a novel adversarial attack model against deep user sequence embedding based classification models, which use the sequence of user posts to generate user embeddings and detect malicious users. In the attack, the adversary generates a new post to fool the classifier. We propose a novel end-to-end Personalized Text Generation Attack model, called PETGEN, that simultaneously reduces the efficacy of the detection model and generates posts that have several key desirable properties. Specifically, PETGEN generates posts that are personalized to the user's writing style, have knowledge about a given target context, are aware of the user's historical posts on the target context, and encapsulate the user's recent topical interests. We conduct extensive experiments on two real-world datasets (Yelp and Wikipedia, both with ground-truth of malicious users) to show that PETGEN significantly reduces the performance of popular deep user sequence embedding-based classification models. PETGEN outperforms five attack baselines in terms of text quality and attack efficacy in both white-box and black-box classifier settings. Overall, this work paves the path towards the next generation of adversary-aware sequence classification models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑