Abstract Cyberbullying has experienced a surge in recent years, mainly due to the widespread adoption of social media platforms. This trend manifests in multiple ways, with hostile language being one of the most common. The latter underscores the urgent need for robust detection methods to address this issue effectively. To address this problem, we propose a novel pipeline to enhance hostile language detection in social media. Our approach consists of a combination of two ideas: First, we propose conducting a Domain Adaptation procedure to specialize the knowledge of a pre-trained BERT, making it more specialized in the domain of social media. For this adaptation, we modify the traditional random Masked Language Modeling technique and propose three novel strategies for selecting the subset of tokens to mask out cleverly. Second, we tailor an Adversarial Regularizer when fine-tuning the adapted BERT for specific hostile-language datasets. We evaluate the performance of our method for detecting hate speech, aggressiveness, offensiveness, and sexism. Our results show that the Domain Adaptation procedure significantly outperforms vanilla BERT, and the Adversarial Regularizer can lead to more robust fine-tuning, thereby enhancing performance. Moreover, we demonstrate that these methods can be used together to achieve an even more significant performance boost.

In recent years, there has been a growing research interest focused on identifying traces of mental disorders through social media analysis. These disorders significantly impair millions of individuals' cognitive and behavioral functions worldwide. Our study aims to advance the understanding of four prevalent mental disorders: Anorexia, Depression, Gambling, and Self-harm. We present a comprehensive framework designed for the domain adaptation of models to analyze and identify signs of these conditions on social media posts. The language models' adapting strategy consisted of three key stages. First, we gathered and enriched substantial data on the four psychological disorders. Second, we adapted the different models to the language used to discuss mental health concerns on social media. Finally, we employed an adapter to fine-tune the models for multiple classification tasks (specific to each mental health condition). The intuitive idea is to adapt a language model smoothly to each domain. Our work includes a comparative study of different language models under in- and cross-domain conditions. This allows us to, for example, assess the ability of a depression-based language model to detect signs of disorders such as anorexia or self-harm. We show that the resulting mental health models perform well in early risk detection tasks. Additionally, we thoroughly analyze the linguistic qualities of these models by testing their predictive abilities using conventional clinical tools, such as specialized questionnaires. We rigorously examine the models across multiple predictive tasks to provide evidence of the adaptation approach's robustness and effectiveness. Our evaluation results are promising. They demonstrate that our framework enhances classification performance and competes favorably with state-of-the-art models.

Emilio Cueva, Adrian Lopez Monroy, Fernando Sánchez-Vega, Thamar Solorio. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Mario Aragon, Adrian Pastor Lopez Monroy, Luis Gonzalez, David E. Losada, Manuel Montes. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Multimodal machine learning is a cutting-edge field that explores ways to incorporate information from multiple sources into models.As more multimodal data becomes available, this field has become increasingly relevant.This work focuses on two key challenges in multimodal machine learning.The first is finding efficient ways to combine information from different data types.The second is that often, one modality (e.g., text) is stronger and more relevant, making it difficult to identify meaningful patterns in the weaker modality (e.g., image).Our approach focuses on more effectively exploiting the weaker modality while dynamically regularizing the loss function.First, we introduce a new two-stream model called Multimodal BERT-ViT, which features a novel intra-CLS token fusion.Second, we utilize a dynamic adjustment that maintains a balance between specialization and generalization during the training to avoid overfitting, which we devised.We add this dynamic adjustment to the Unsupervised Data Augmentation (UDA) framework.We evaluate the effectiveness of these proposals on the task of multi-label movie genre classification using the Moviescope and MM-IMDb datasets.The evaluation revealed that our proposal offers substantial benefits, while simultaneously enabling us to harness the weaker modality without compromising the information provided by the stronger.

Millions of people around the world are affected by one or more mental disorders that interfere in their thinking and behavior. A timely detection of these issues is challenging but crucial, since it could open the possibility to offer help to people before the illness gets worse. One alternative to accomplish this is to monitor how people express themselves, that is for example what and how they write, or even a step further, what emotions they express in their social media communications. In this article, we analyze two computational representations that aim to model the presence and changes of the emotions expressed by social media users. In our evaluation we use two recent public data sets for two important mental disorders: Depression and Anorexia. The obtained results suggest that the presence and variability of emotions, captured by the proposed representations, allow to highlight important information about social media users suffering from depression or anorexia. Furthermore, the fusion of both representations can boost the performance, equalling the best reported approach for depression and barely behind the top performer for anorexia by only 1 percent. Moreover, these representations open the possibility to add some interpretability to the results.

Gustavo Aguilar, Adrian Pastor López-Monroy, Fabio González, Thamar Solorio. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers). 2018.

Adrian Pastor López-Monroy, Fabio A. González, Manuel Montes, Hugo Jair Escalante, Thamar Solorio. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers). 2018.