Трофімов O., & Hulak, H. (2026). Method for enhancing the robustness of large language models against adversarial attacks. Ukrainian Scientific Journal of Information Security, 32(2), 72–82. https://doi.org/10.18372/2225-5036.32.21660