Digital communication channels have become major vectors for large-scale cyber fraud, including spam, phishing, and recruitment scams, causing significant financial losses and eroding user trust. While transformer-based models have improved text classification, most systems remain English-centric, limiting their effectiveness in multilingual environments and across diverse fraud typologies. In this work, we hypothesize that fraud detection models trained in a single language can exhibit cross-lingual generalization capabilities. We investigate the use of Natural Language Processing (NLP) for detecting multiple forms of textual fraud by developing and evaluating three transformer-based discriminative models targeting spam, phishing, and fake job postings. Specifically, we compare two architectures—MrBERT (308M parameters) and mROBERTa (283M parameters)—across English, Spanish, and Valencian data. Our results demonstrate that while models trained exclusively on English achieve near-perfect performance in-language (F1 ≈ 0.98), they exhibit limited cross-lingual generalization in other languages such as Spanish (F1 ≈ 0.44 for phishing detection, and F1 ≈ 0.85 for spam detection). However, incorporating multilingual training data dramatically improves performance in target languages (F1 ≈ 0.95–0.99) while maintaining or even enhancing English accuracy. We further observe that cross-lingual transfer is most effective when datasets are parallel or closely aligned. These findings underscore the critical importance of multilingual data inclusion for building robust, scalable cybersecurity frameworks in diverse linguistic settings.

This paper describes our approach to address the SemEval-2025 Task 10 subtask 3, which is focused on narrative extraction given news articles with a dominant narrative. We design an external knowledge injection approach to fine-tune a Flan-T5 model so the generated narrative explanations are in line with the dominant narrative determined in each text. We also incorporate pragmatic information in the form of communicative intentions, using them as external knowledge to assist the model. This ensures that the generated texts align more closely with the intended explanations and effectively convey the expected meaning. The results show that our approach ranks 3rd in the task leaderboard (0.7428 in Macro-F1) with concise and effective news explanations. The analyses highlight the importance of adding pragmatic information when training systems to generate adequate narrative extractions.

This paper addresses a key limitation in Natural Language Generation (NLG) systems: their struggle with commonsense reasoning, which is essential for generating contextually appropriate and plausible text. The study proposes an approach to enhance the commonsense reasoning abilities of NLG systems by integrating external knowledge framed in a constrained commonsense generation task. The paper investigates strategies for extracting and injecting external knowledge into pre-trained models, specifically BART and T5, in both base and large configurations. Experimental results show that incorporating external knowledge extracted with a simple strategy leads to significant improvements in performance, with the models achieving 88% accuracy in generating plausible and correct sentences. When refined methods for knowledge extraction are applied, the accuracy further increases to 92%. These findings underscore the crucial role of high-quality external knowledge in enhancing the commonsense reasoning capabilities of NLG systems, suggesting that such integration is vital for advancing their performance in real-world applications.