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 2303.05077cs.CLcs.CV
收藏

学习视觉文本扰动的易读性

Learning the Legibility of Visual Text Perturbations

Carnegie Mellon University(卡内基梅隆大学);Duke University(杜克大学)

Dev Seth, Rickard Stureborg, Danish Pruthi, Bhuwan Dhingra

摘要:

NLP中的许多对抗攻击会干扰输入以产生视觉上相似的字符串('ergo' $\rightarrow$ ' $ε$ rgo'),这些字符串对人类来说是清晰的,但会降低模型性能。虽然保持易读性是文本扰动的必要条件,但很少有系统地描述它的工作;相反,易读性通常是通过对扰动的性质和程度的直觉来松散地执行的。特别是,目前还不清楚在多大程度上可以扰动输入,同时保持易读性,或如何量化的易读性扰动字符串。在这项工作中,我们通过学习模型来解决这个差距,这些模型预测扰动字符串的易读性,并根据其易读性对候选扰动进行排名。为此,我们收集并发布了LEGIT,这是一个人工注释的数据集,包括视觉干扰文本的易读性。使用这个数据集,我们建立了基于文本和基于视觉的模型,在预测输入是否清晰方面达到了$0.91$ F1分数,并且在预测两个给定扰动中哪一个更清晰方面达到了$0.86$的准确性。此外,我们发现,来自LEGIT数据集的清晰扰动在降低NLP模型的性能方面比最知名的攻击策略更有效,这表明当前模型可能容易受到现有视觉攻击所捕获的各种扰动的影响。数据、代码和模型可在https://github.com/dvsth/learning-legibility-2023上获得。

英文摘要:

Many adversarial attacks in NLP perturb inputs to produce visually similar strings ('ergo' $\rightarrow$ '$ε$rgo') which are legible to humans but degrade model performance. Although preserving legibility is a necessary condition for text perturbation, little work has been done to systematically characterize it; instead, legibility is typically loosely enforced via intuitions around the nature and extent of perturbations. Particularly, it is unclear to what extent can inputs be perturbed while preserving legibility, or how to quantify the legibility of a perturbed string. In this work, we address this gap by learning models that predict the legibility of a perturbed string, and rank candidate perturbations based on their legibility. To do so, we collect and release LEGIT, a human-annotated dataset comprising the legibility of visually perturbed text. Using this dataset, we build both text- and vision-based models which achieve up to $0.91$ F1 score in predicting whether an input is legible, and an accuracy of $0.86$ in predicting which of two given perturbations is more legible. Additionally, we discover that legible perturbations from the LEGIT dataset are more effective at lowering the performance of NLP models than best-known attack strategies, suggesting that current models may be vulnerable to a broad range of perturbations beyond what is captured by existing visual attacks. Data, code, and models are available at https://github.com/dvsth/learning-legibility-2023.

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。

↑