An image is worth a thousand toxic words: A metamorphic testing framework for content moderation software

W Wang, J Huang, J Huang, C Chen… - 2023 38th IEEE/ACM …, 2023 - ieeexplore.ieee.org
The exponential growth of social media platforms has brought about a revolution in
communication and content dissemination in human society. Nevertheless, these platforms …

GReAT: A graph regularized adversarial training method

S Bayram, K Barner - IEEE Access, 2024 - ieeexplore.ieee.org
This paper presents GReAT (Graph Regularized Adversarial Training), a novel
regularization method designed to enhance the robust classification performance of deep …

Test Time Augmentation as a Defense Against Adversarial Attacks on Online Handwriting

Y Yamashita, BK Iwana - International Conference on Document Analysis …, 2024 - Springer
Neural networks have been shown to be weak against adversarial attacks. This study
examines the effects of adversarial attacks on online handwritten characters and proposes a …

From Characters to Chaos: On the Feasibility of Attacking Thai OCR with Adversarial Examples

C Jiamsuchon, J Suaboot… - 2023 20th International …, 2023 - ieeexplore.ieee.org
Recent advances in deep neural networks (DNNs) have significantly enhanced the
capabilities of optical character recognition (OCR) technology, enabling its adoption to a …

On the feasibility of attacking Thai LPR systems with adversarial examples

C Jiamsuchon, J Suaboot… - arxiv preprint arxiv …, 2023 - arxiv.org
Recent advances in deep neural networks (DNNs) have significantly enhanced the
capabilities of optical character recognition (OCR) technology, enabling its adoption to a …