The rise of social media has amplified the spread of fake news, threatening societal stability. To address this, machine learning-based fake news detection systems analyze news content, including text, images, and user interactions. However, these systems are vulnerable to attacks that manipulate data, undermining their effectiveness. Studying these vulnerabilities is essential for improving fake news detectors. While much research has focused on attacks targeting news articles, less attention has been given to user comments, which play a crucial role in how news is perceived. Existing attack methods either reuse generic comments or generate new ones, but these approaches often fail to generate comments relevant or realistic enough to deceive detection systems. In this article, we focus on generating more realistic comments for specific news articles. We propose three novel attack strategies: two fine-tune a small language model using reinforcement learning algorithms transformer reinforcement learning and reinforced self-training, while the third uses a large language model with self-reflection. We conduct experiments showing that our methods outperform the state-of-the-art MALCOM procedure and that our proposed attack strategies produce more realistic comments, which are consequently more difficult to detect.

Adversarial attacks pose significant challenges to deep neural networks (DNNs) such as Transformer models in natural language processing (NLP). This article introduces a novel defense strategy, called GenFighter , which enhances adversarial robustness by learning and reasoning on the training classification distribution. GenFighter identifies potentially malicious instances deviating from the distribution, transforms them into semantically equivalent instances aligned with the training data, and employs ensemble techniques for a unified and robust response. By conducting extensive experiments, we show that GenFighter outperforms state-of-the-art defenses in accuracy under attack and attack success rate metrics while maintaining the same or superior generalization capabilities. Additionally, it requires a high number of queries per attack, making the attack more challenging in real scenarios. Finally, The ablation study shows that our approach proficiently integrates transfer learning, a generative/evolutive procedure, and an ensemble method, providing an effective defense against NLP adversarial attacks.

Context: Agile development methodologies in the software industry have increased significantly over the past decade. Although one of the main aspects of agile software development (ASD) is less documentation, there have always been conflicting opinions about what to document in ASD. Objective: This study aims to systematically identify what to document in ASD, which documentation tools and methods are in use, and how those tools can overcome documentation challenges. Method: We performed a systematic literature review of the studies published between 2010 and June 2021 that discusses agile documentation. Then, we systematically selected a pool of 74 studies using particular inclusion and exclusion criteria. After that, we conducted a quantitative and qualitative analysis using the data extracted from these studies. Results: We found nine primary vital factors to add to agile documentation from our pool of studies. Our analysis shows that agile practitioners have primarily developed their documentation tools and methods focusing on these factors. The results suggest that the tools and techniques in agile documentation are not in sync, and they separately solve different challenges. Conclusions: Based on our results and discussion, researchers and practitioners will better understand how current agile documentation tools and practices perform. In addition, investigation of the synchronization of these tools will be helpful in future research and development.