Addressing the need for effective hate speech moderation in contemporary digital discourse, the Multimodal Hate Speech Event Detection Shared Task made its debut at CASE 2023, co-located with RANLP 2023. Building upon its success, an extended version of the shared task was organized at the CASE workshop in EACL 2024. Similar to the earlier iteration, in this shared task, participants address hate speech detection through two subtasks. Subtask A is a binary classification problem, assessing whether text-embedded images contain hate speech. Subtask B goes further, demanding the identification of hate speech targets, such as individuals, communities, and organizations within text-embedded images. Performance is evaluated using the macro F1-score metric in both subtasks. With a total of 73 registered participants, the shared task witnessed remarkable achievements, with the best F1-scores in Subtask A and Subtask B reaching 87.27% and 80.05%, respectively, surpassing the leaderboard of the previous CASE 2023 shared task. This paper provides a comprehensive overview of the performance of seven teams that submitted results for Subtask A and five teams for Subtask B.

Social media plays a pivotal role in global discussions, including on climate change. The variety of opinions expressed range from supportive to oppositional, with some instances of hate speech. Recognizing the importance of understanding these varied perspectives, the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE) at EACL 2024 hosted a shared task focused on detecting stances and hate speech in climate activism-related tweets. This task was divided into three subtasks: subtasks A and B concentrated on identifying hate speech and its targets, while subtask C focused on stance detection. Participants’ performance was evaluated using the macro F1-score. With over 100 teams participating, the highest F1 scores achieved were 91.44% in subtask C, 78.58% in subtask B, and 74.83% in subtask A. This paper details the methodologies of 24 teams that submitted their results to the competition’s leaderboard.

Ensuring the moderation of hate speech and its targets emerges as a critical imperative within contemporary digital discourse. To facilitate this imperative, the shared task Multimodal Hate Speech Event Detection was organized in the sixth CASE workshop co-located at RANLP 2023. The shared task has two subtasks. The sub-task A required participants to pose hate speech detection as a binary problem i.e. they had to detect if the given text-embedded image had hate or not. Similarly, sub-task B required participants to identify the targets of the hate speech namely individual, community, and organization targets in text-embedded images. For both sub-tasks, the participants were ranked on the basis of the F1-score. The best F1-score in sub-task A and sub-task B were 85.65 and 76.34 respectively. This paper provides a comprehensive overview of the performance of 13 teams that submitted the results in Subtask A and 10 teams in Subtask B.

Natural language processing has advanced with AI-driven language models (LMs), that are applied widely from text generation to question answering. These models are pre-trained on a wide spectrum of data sources, enhancing accuracy and responsiveness. However, this process inadvertently entails the absorption of a diverse spectrum of viewpoints inherent within the training data. Exploring political leaning within LMs due to such viewpoints remains a less-explored domain. In the context of a low-resource language like Bangla, this area of research is nearly non-existent. To bridge this gap, we comprehensively analyze biases present in Bangla language models, specifically focusing on social and economic dimensions. Our findings reveal the inclinations of various LMs, which will provide insights into ethical considerations and limitations associated with deploying Bangla LMs.

Violence incitement detection and sentiment analysis hold significant importance in the field of natural language processing. However, in the case of the Bangla language, there are unique challenges due to its low-resource nature. In this paper, we address these challenges by presenting an innovative approach that leverages aggregated BERT models for two tasks at the BLP workshop in EMNLP 2023, specifically tailored for Bangla. Task 1 focuses on violence-inciting text detection, while task 2 centers on sentiment analysis. Our approach combines fine-tuning with textual entailment (utilizing BanglaBERT), Masked Language Model (MLM) training (making use of BanglaBERT), and the use of standalone Multilingual BERT. This comprehensive framework significantly enhances the accuracy of sentiment classification and violence incitement detection in Bangla text. Our method achieved the 11th rank in task 1 with an F1-score of 73.47 and the 4th rank in task 2 with an F1-score of 71.73. This paper provides a detailed system description along with an analysis of the impact of each component of our framework.

Alzheimer’s Disease (AD) is a neurodegenerative disorder that affects cognitive abilities and memory, especially in older adults. One of the challenges of AD is that it can be difficult to diagnose in its early stages. However, recent research has shown that changes in language, including speech decline and difficulty in processing information, can be important indicators of AD and may help with early detection. Hence, the speech narratives of the patients can be useful in diagnosing the early stages of Alzheimer’s disease. While the previous works have presented the potential of using speech narratives to diagnose AD in high-resource languages, this work explores the possibility of using a low-resourced language, i.e., Hindi language, to diagnose AD. In this paper, we present a dataset specifically for analyzing AD in the Hindi language, along with experimental results using various state-of-the-art algorithms to assess the diagnostic potential of speech narratives in Hindi. Our analysis suggests that speech narratives in the Hindi language have the potential to aid in the diagnosis of AD. Our dataset and code are made publicly available at https://github.com/rkritesh210/DementiaBankHindi.

Graph-based techniques have gained traction for representing and analyzing data in various natural language processing (NLP) tasks. Knowledge graph-based language representation models have shown promising results in leveraging domain-specific knowledge for NLP tasks, particularly in the biomedical NLP field. However, such models have limitations, including knowledge noise and neglect of contextual relationships, leading to potential semantic errors and reduced accuracy. To address these issues, this paper proposes two novel methods. The first method combines knowledge graph-based language model with nearest-neighbor models to incorporate semantic and category information from neighboring instances. The second method involves integrating knowledge graph-based language model with graph neural networks (GNNs) to leverage feature information from neighboring nodes in the graph. Experiments on relation extraction (RE) and classification tasks in English and Chinese language datasets demonstrate significant performance improvements with both methods, highlighting their potential for enhancing the performance of language models and improving NLP applications in the biomedical domain.

Despite the remarkable performances in various applications, machine learning (ML) models could potentially discriminate. They may result in biasness in decision-making, leading to an impact negatively on individuals and society. Recently, various methods have been developed to mitigate biasness and achieve significant performance. Attention mechanisms are a fundamental component of many state-of-the-art ML models and may potentially impact the fairness of ML models. However, how they explicitly influence fairness has yet to be thoroughly explored. In this paper, we investigate how different attention mechanisms affect the fairness of ML models, focusing on models used in Natural Language Processing (NLP) models. We evaluate the performance of fairness of several models with and without different attention mechanisms on widely used benchmark datasets. Our results indicate that the majority of attention mechanisms that have been assessed can improve the fairness performance of Bidirectional Gated Recurrent Unit (BiGRU) and Bidirectional Long Short-Term Memory (BiLSTM) in all three datasets regarding religious and gender-sensitive groups, however, with varying degrees of trade-offs in accuracy measures. Our findings highlight the possibility of fairness being affected by adopting specific attention mechanisms in machine learning models for certain datasets

In the evolving landscape of Environmental, Social, and Corporate Governance (ESG) impact assessment, the ML-ESG-2 shared task proposes identifying ESG impact types. To address this challenge, we present a comprehensive system leveraging ensemble learning techniques, capitalizing on early and late fusion approaches. Our approach employs four distinct models: mBERT, FlauBERT-base, ALBERT-base-v2, and a Multi-Layer Perceptron (MLP) incorporating Latent Semantic Analysis (LSA) and Term Frequency-Inverse Document Frequency (TF-IDF) features. Through extensive experimentation, we find that our early fusion ensemble approach, featuring the integration of LSA, TF-IDF, mBERT, FlauBERT-base, and ALBERT-base-v2, delivers the best performance. Our system offers a comprehensive ESG impact type identification solution, contributing to the responsible and sustainable decision-making processes vital in today’s financial and corporate governance landscape.

In an era of widespread digital communication, the challenge of identifying and countering disinformation has become increasingly critical. However, compared to the solutions available in the English language, the resources and strategies for tackling this multifaceted problem in Arabic are relatively scarce. To address this issue, this paper presents our solutions to tasks in ArAIEval 2023. Task 1 focuses on detecting persuasion techniques, while Task 2 centers on disinformation detection within Arabic text. Leveraging a multi-head model architecture, fine-tuning techniques, sequential learning, and innovative activation functions, our contributions significantly enhance persuasion techniques and disinformation detection accuracy. Beyond improving performance, our work fills a critical research gap in content analysis for Arabic, empowering individuals, communities, and digital platforms to combat deceptive content effectively and preserve the credibility of information sources within the Arabic-speaking world.

With approximately 400 million speakers worldwide, Arabic ranks as the fifth most-spoken language globally, necessitating advancements in natural language processing. This paper addresses this need by presenting a system description of the approaches employed for the subtasks outlined in the Nuanced Arabic Dialect Identification (NADI) task at EMNLP 2023. For the first subtask, involving closed country-level dialect identification classification, we employ an ensemble of two Arabic language models. Similarly, for the second subtask, focused on closed dialect to Modern Standard Arabic (MSA) machine translation, our approach combines sequence-to-sequence models, all trained on an Arabic-specific dataset. Our team ranks 10th and 3rd on subtask 1 and subtask 2 respectively.

The Qur’an holds immense theological and historical significance, and developing a technology-driven solution for answering questions from this sacred text is of paramount importance. This paper presents our approach to task B of Qur’an QA 2023, part of EMNLP 2023, addressing this challenge by proposing a robust method for extracting answers from Qur’anic passages. Leveraging the Qur’anic Reading Comprehension Dataset (QRCD) v1.2, we employ innovative techniques and advanced models to improve the precision and contextuality of answers derived from Qur’anic passages. Our methodology encompasses the utilization of start and end logits, Long Short-Term Memory (LSTM) networks, and fusion mechanisms, contributing to the ongoing dialogue at the intersection of technology and spirituality.

Abstractive related work generation has attracted increasing attention in generating coherent related work that helps readers grasp the current research. However, most existing models ignore the inherent causality during related work generation, leading to spurious correlations which downgrade the models’ generation quality and generalizability. In this study, we argue that causal intervention can address such limitations and improve the quality and coherence of generated related work. To this end, we propose a novel Causal Intervention Module for Related Work Generation (CaM) to effectively capture causalities in the generation process. Specifically, we first model the relations among the sentence order, document (reference) correlations, and transitional content in related work generation using a causal graph. Then, to implement causal interventions and mitigate the negative impact of spurious correlations, we use do-calculus to derive ordinary conditional probabilities and identify causal effects through CaM. Finally, we subtly fuse CaM with Transformer to obtain an end-to-end related work generation framework. Extensive experiments on two real-world datasets show that CaM can effectively promote the model to learn causal relations and thus produce related work of higher quality and coherence.

Clickbait posts tend to spread inaccurate or misleading information to manipulate people’s attention and emotions, which greatly harms the credibility of social media. Existing clickbait detection models rely on analyzing the objective semantics in posts or correlating posts with article content only. However, these models fail to identify and exploit the manipulation intention of clickbait from a user’s subjective perspective, leading to limited capability to explore comprehensive clues of clickbait. To address such a issue, we propose a multiview clickbait detection model, named MCDM, to model subjective and objective preferences simultaneously. MCDM introduces two novel complementary modules for modeling subjective feeling and objective content relevance, respectively. The subjective feeling module adopts a user-centric approach to capture subjective features of posts, such as language patterns and emotional inclinations. The objective module explores news elements from posts and models article content correlations to capture objective clues for clickbait detection. Extensive experimental results on two real-world datasets show that our proposed MCDM outperforms state-of-the-art approaches for clickbait detection, verifying the effectiveness of integrating subjective and objective preferences for detecting clickbait.

Medical dialogue systems have the potential to assist doctors in expanding access to medical care, improving the quality of patient experiences, and lowering medical expenses. The computational methods are still in their early stages and are not ready for widespread application despite their great potential. Existing transformer-based language models have shown promising results but lack domain-specific knowledge. However, to diagnose like doctors, an automatic medical diagnosis necessitates more stringent requirements for the rationality of the dialogue in the context of relevant knowledge. In this study, we propose a new method that addresses the challenges of medical dialogue generation by incorporating medical knowledge into transformer-based language models. We present a method that leverages an external medical knowledge graph and injects triples as domain knowledge into the utterances. Automatic and human evaluation on a publicly available dataset demonstrates that incorporating medical knowledge outperforms several state-of-the-art baseline methods.

A user-generated text on social media enables health workers to keep track of information, identify possible outbreaks, forecast disease trends, monitor emergency cases, and ascertain disease awareness and response to official health correspondence. This exchange of health information on social media has been regarded as an attempt to enhance public health surveillance (PHS). Despite its potential, the technology is still in its early stages and is not ready for widespread application. Advancements in pretrained language models (PLMs) have facilitated the development of several domain-specific PLMs and a variety of downstream applications. However, there are no PLMs for social media tasks involving PHS. We present and release PHS-BERT, a transformer-based PLM, to identify tasks related to public health surveillance on social media. We compared and benchmarked the performance of PHS-BERT on 25 datasets from different social medial platforms related to 7 different PHS tasks. Compared with existing PLMs that are mainly evaluated on limited tasks, PHS-BERT achieved state-of-the-art performance on all 25 tested datasets, showing that our PLM is robust and generalizable in the common PHS tasks. By making PHS-BERT available, we aim to facilitate the community to reduce the computational cost and introduce new baselines for future works across various PHS-related tasks.

News recommender systems face certain challenges. These challenges arise due to evolving users’ preferences over dynamically created news articles. The diversity is necessary for a news recommender system to expose users to a variety of information. We propose a deep neural network based on a two-tower architecture that learns news representation through a news item tower and users’ representations through a query tower. We customize an augmented vector for each query and news item to introduce information interaction between the two towers. We introduce diversity in the proposed architecture by considering a category loss function that aligns items’ representation of uneven news categories. Experimental results on two news datasets reveal that our proposed architecture is more effective compared to the state-of-the-art methods and achieves a balance between accuracy and diversity.

This paper presents a new multi-modal dataset for identifying hateful content on social media, consisting of 5,680 text-image pairs collected from Twitter, labeled across two labels. Experimental analysis of the presented dataset has shown that understanding both modalities is essential for detecting these techniques. It is confirmed in our experiments with several state-of-the-art multi-modal models. In future work, we plan to extend the dataset in size. We further plan to develop new multi-modal models tailored explicitly to hate-speech detection, aiming for a deeper understanding of the text and image relation. It would also be interesting to perform experiments in a direction that explores what social entities the given hate speech tweet targets.