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.

Mental disorders affect millions of people worldwide and cause interference with their thinking and behavior. Through the past years, awareness created by health campaigns and other sources motivated the study of these disorders using information extracted from social media platforms. In this work, we aim to contribute to the study of these disorders and to the understanding of how mental problems reflect on social media. To achieve this goal, we propose a double-domain adaptation of a language model. First, we adapted the model to social media language, and then, we adapted it to the mental health domain. In both steps, we incorporated a lexical resource to guide the masking process of the language model and, therefore, to help it in paying more attention to words related to mental disorders. We have evaluated our model in the detection of signs of three major mental disorders: Anorexia, Self-harm, and Depression. Results are encouraging as they show that the proposed adaptation enhances the classification performance and yields competitive results against state-of-the-art methods.

This paper describes our participation in the Clickbait challenge at SemEval 2023. In this work, we address the Clickbait classification task using transformers models in an ensemble configuration. We tackle the Spoiler Generation task using a two-level ensemble strategy of models trained for extractive QA, and selecting the best K candidates for multi-part spoilers. In the test partitions, our approaches obtained a classification accuracy of 0.716 for classification and a BLEU-4 score of 0.439 for spoiler generation.

The Multimodal Transformer showed to be a competitive model for multimodal tasks involving textual, visual and audio signals. However, as more modalities are involved, its late fusion by concatenation starts to have a negative impact on the model’s performance. Besides, interpreting model’s predictions becomes difficult, as one would have to look at the different attention activation matrices. In order to overcome these shortcomings, we propose to perform late fusion by adding a GMU module, which effectively allows the model to weight modalities at instance level, improving its performance while providing a better interpretabilty mechanism. In the experiments, we compare our proposed model (MulT-GMU) against the original implementation (MulT-Concat) and a SOTA model tested in a movie genre classification dataset. Our approach, MulT-GMU, outperforms both, MulT-Concat and previous SOTA model.

Nowadays, the amount of users’ activities on online social media is growing dramatically. These online environments provide excellent opportunities for communication and knowledge sharing. However, some people misuse them to harass and bully others online, a phenomenon called cyberbullying. Due to its harmful effects on people, especially youth, it is imperative to detect cyberbullying as early as possible before it causes irreparable damages to victims. Most of the relevant available resources are not explicitly designed to detect cyberbullying, but related content, such as hate speech and abusive language. In this paper, we propose a new approach to create a corpus suited for cyberbullying detection. We also investigate the possibility of designing a framework to monitor the streams of users’ online messages and detects the signs of cyberbullying as early as possible.

Nowadays social media platforms are the most popular way for people to share information, from work issues to personal matters. For example, people with health disorders tend to share their concerns for advice, support or simply to relieve suffering. This provides a great opportunity to proactively detect these users and refer them as soon as possible to professional help. We propose a new representation called Bag of Sub-Emotions (BoSE), which represents social media documents by a set of fine-grained emotions automatically generated using a lexical resource of emotions and subword embeddings. The proposed representation is evaluated in the task of depression detection. The results are encouraging; the usage of fine-grained emotions improved the results from a representation based on the core emotions and obtained competitive results in comparison to state of the art approaches.

The intensive use of e-communications in everyday life has given rise to new threats and risks. When the vulnerable asset is the user, detecting these potential attacks before they cause serious damages is extremely important. This paper proposes a novel document representation to improve the early detection of risks in social media sources. The goal is to effectively identify the potential risk using as few text as possible and with as much anticipation as possible. Accordingly, we devise a Multi-Resolution Representation (MulR), which allows us to generate multiple “views” of the analyzed text. These views capture different semantic meanings for words and documents at different levels of detail, which is very useful in early scenarios to model the variable amounts of evidence. Intuitively, the representation captures better the content of short documents (very early stages) in low resolutions, whereas large documents (medium/large stages) are better modeled with higher resolutions. We evaluate the proposed ideas in two different tasks where anticipation is critical: sexual predator detection and depression detection. The experimental evaluation for these early tasks revealed that the proposed approach outperforms previous methodologies by a considerable margin.

Recognizing named entities in a document is a key task in many NLP applications. Although current state-of-the-art approaches to this task reach a high performance on clean text (e.g. newswire genres), those algorithms dramatically degrade when they are moved to noisy environments such as social media domains. We present two systems that address the challenges of processing social media data using character-level phonetics and phonology, word embeddings, and Part-of-Speech tags as features. The first model is a multitask end-to-end Bidirectional Long Short-Term Memory (BLSTM)-Conditional Random Field (CRF) network whose output layer contains two CRF classifiers. The second model uses a multitask BLSTM network as feature extractor that transfers the learning to a CRF classifier for the final prediction. Our systems outperform the current F1 scores of the state of the art on the Workshop on Noisy User-generated Text 2017 dataset by 2.45% and 3.69%, establishing a more suitable approach for social media environments.

Named Entity Recognition for social media data is challenging because of its inherent noisiness. In addition to improper grammatical structures, it contains spelling inconsistencies and numerous informal abbreviations. We propose a novel multi-task approach by employing a more general secondary task of Named Entity (NE) segmentation together with the primary task of fine-grained NE categorization. The multi-task neural network architecture learns higher order feature representations from word and character sequences along with basic Part-of-Speech tags and gazetteer information. This neural network acts as a feature extractor to feed a Conditional Random Fields classifier. We were able to obtain the first position in the 3rd Workshop on Noisy User-generated Text (WNUT-2017) with a 41.86% entity F1-score and a 40.24% surface F1-score.