Exploiting Explainability to Design Adversarial Attacks and Evaluate Attack Resilience in Hate-Speech Detection Models
Year: 2025
Type: conference-paper
Abstract: The advent of social media has given rise to numerous ethical challenges, with hate speech among the most significant concerns. Researchers are attempting to tackle this problem by using hate-speech detection and employing language models to automatically moderate content and promote civil discourse. Unfortunately, recent studies have revealed that hate-speech detection systems can be misled by adversarial attacks, raising concerns about their resilience. While previous research has separately addressed the robustness of these models under adversarial attacks and their explainability, there has been no comprehensive study exploring their intersection. The novelty of our work lies in combining these two critical aspects, leveraging explainability to identify potential vulnerabilities and enabling the design of targeted adversarial attacks. This paper quantifies the interplay between explainability and adversarial robustness in hate-speech detection models. We define novel metrics based on explainability-driven adversarial attacks to evaluate this relationship, providing a clear assessment of model vulnerabilities and guiding the development of more resilient systems.
Authors: Pranath Reddy Kumbam, Sohaib Uddin Syed, Prashanth Thamminedi, Suhas Harish, Ian Perera, Bonnie J Dorr
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4411120689