VertAttack: Taking advantage of Text Classifiers' horizontal vision
Text classification systems have continuously improved in performance over the years. However, nearly all current SOTA classifiers have a similar shortcoming, they process text in a horizontal manner. Vertically written words will not be recognized by a classifier. In contrast, humans are easily able to recognize and read words written both horizontally and vertically. Hence, a human adversary could write problematic words vertically and the meaning would still be preserved to other humans. We simulate such an attack, VertAttack. VertAttack identifies which words a classifier is reliant on and then rewrites those words vertically. We find that VertAttack is able to greatly drop the accuracy of 4 different transformer models on 5 datasets. For example, on the SST2 dataset, VertAttack is able to drop RoBERTa's accuracy from 94 to 13%. Furthermore, since VertAttack does not replace the word, meaning is easily preserved. We verify this via a human study and find that crowdworkers are able to correctly label 77% perturbed texts perturbed, compared to 81% of the original texts. We believe VertAttack offers a look into how humans might circumvent classifiers in the future and thus inspire a look into more robust algorithms.
Code (1)
Tasks
text-classificationText ClassificationSimilar Papers 제목 키워드 기반
Continual Horizontal Federated Learning for Heterogeneous Data
Federated learning is a promising machine learning technique that enables multiple clients to collaboratively build a model without revealing the raw data to each other. Among various types of federated learning methods,…
Continual LearningFederated LearningSingle-Stage Rotation-Decoupled Detector for Oriented Object
Oriented object detection has received extensive attention in recent years, especially for the task of detecting targets in aerial imagery. Traditional detectors locate objects by horizontal bounding boxes (HBBs), which …
Objectobject-detectionObject DetectionObject Detection In Aerial Images+1Don’t sweat the small stuff, classify the rest: Sample Shielding to protect text classifiers against adversarial attacks
Deep learning (DL) is being used extensively for text classification. However, researchers have demonstrated the vulnerability of such classifiers to adversarial attacks. Attackers modify the text in a way which misleads…
text-classificationText ClassificationDon't sweat the small stuff, classify the rest: Sample Shielding to protect text classifiers against adversarial attacks
Deep learning (DL) is being used extensively for text classification. However, researchers have demonstrated the vulnerability of such classifiers to adversarial attacks. Attackers modify the text in a way which misleads…
text-classificationText ClassificationA preprocessing perspective for quantum machine learning classification advantage using NISQ algorithms
Quantum Machine Learning (QML) hasn't yet demonstrated extensively and clearly its advantages compared to the classical machine learning approach. So far, there are only specific cases where some quantum-inspired techniq…
Dimensionality ReductionQuantum Machine Learning