The rapid expansion of social media platforms has led to a significant increase in the spread of harmful content, including misogynistic, homophobic, and transphobic memes. Detecting such content is challenging because memes often combine textual and visual elements and frequently appear in multilingual and culturally diverse contexts. This study proposes a multimodal transformer-based framework for multilingual harmful meme classification that integrates textual and visual representations to improve detection performance. The proposed architecture employs XLM-RoBERTa for multilingual text encoding and the Swin Transformer for hierarchical visual feature extraction. A cross-attention fusion mechanism is introduced to enable meaningful interaction between textual and visual modalities. The fused representation is then processed through a classification layer to perform multi-class prediction. Experiments are conducted across multiple datasets covering eight languages and three harmful content categories: misogyny, homophobia/transphobia, and hate speech. The model is evaluated using the macro-F1 score and demonstrates consistent improvements over baseline multimodal systems across both high-resource and low-resource languages. The results highlight the effectiveness of transformer-based multimodal architectures in capturing implicit and contextual harmful signals present in memes. The study contributes to the development of robust multilingual systems for harmful content detection and supports efforts toward creating safer and more inclusive online environments.

This paper presents an overview of the second shared task on Abusive Tamil Text Targeting Women on Social Media as a binary classification problem (abusive vs. non-abusive). We release a dataset of Tamil YouTube comments and evaluate submissions using macro-F1 to encourage balanced performance in a noisy, low-resource setting. There are 89 teams registered for this task and 24 teams submitted the results. The approaches used by the teams includes transformer fine-tuning, heterogeneous ensembles, classical baselines, and large language models using prompting and LoRA. Results show that the best-performing system scored 0.8297 macro-F1 and many submissions are around 0.79-0.81. Across submissions, transformer fine-tuning with domain-aligned encoders is consistently strong, while additional gains are frequently associated with Tamil-aware normalization and macro-F1-oriented calibration such as class-weighted learning and validation-based threshold tuning. Overall, the findings highlights the importance of language-aware preprocessing and careful decision calibration for reliable moderation of women-targeted abusive Tamil social media text.Disclaimer: This paper (including figures and examples) may contain offensive or harmful language, including abusive content targeting women. All such text is presented solely for research and educational purposes and it does not reflect the author’s views. Reader discretion is advised.

Hate speech targeting caste and migration communities is a growing concern in online platforms, particularly in linguistically diverse regions. By focusing on Tamil language text content, this task provides a unique opportunity to tackle caste or migration related hate speech detection in a low resource language Tamil, contributing to a safer digital space. We present the results and main findings of the shared task caste and migration hate speech detection. The task is a binary classification determining whether a text is caste/migration related hate speech or not. The task attracted 17 participating teams, experimenting with a wide range of methodologies from traditional machine learning to advanced multilingual transformers. The top performing system achieved a macro F1-score of 0.88105, enhancing an ensemble of fine-tuned transformer models including XLM-R and MuRIL. Our analysis highlights the effectiveness of multilingual transformers in low resource, ensemble learning, and culturally informed socio political context based techniques.

Sentiment analysis is an essential task for interpreting subjective opinions and emotions in textual data, with significant implications across commercial and societal applications. This paper provides an overview of the shared task on Sentiment Analysis in Tamil and Tulu, organized as part of DravidianLangTech@NAACL 2025. The task comprises two components: one addressing Tamil and the other focusing on Tulu, both designed as multi-class classification challenges, wherein the sentiment of a given text must be categorized as positive, negative, neutral and unknown. The dataset was diligently organized by aggregating user-generated content from social media platforms such as YouTube and Twitter, ensuring linguistic diversity and real-world applicability. Participants applied a variety of computational approaches, ranging from classical machine learning algorithms such as Traditional Machine Learning Models, Deep Learning Models, Pre-trained Language Models and other Feature Representation Techniques to tackle the challenges posed by linguistic code-mixing, orthographic variations, and resource scarcity in these low resource languages.

We present an overview of the first shared task on “Caste and Migration Hate Speech Detection.” The shared task is organized as part of LTEDI@EACL 2024. The system must delineate between binary outcomes, ascertaining whether the text is categorized as a caste/migration hate speech or not. The dataset presented in this shared task is in Tamil, which is one of the under-resource languages. There are a total of 51 teams participated in this task. Among them, 15 teams submitted their research results for the task. To the best of our knowledge, this is the first time the shared task has been conducted on textual hate speech detection concerning caste and migration. In this study, we have conducted a systematic analysis and detailed presentation of all the contributions of the participants as well as the statistics of the dataset, which is the social media comments in Tamil language to detect hate speech. It also further goes into the details of a comprehensive analysis of the participants’ methodology and their findings.

This paper offers a detailed overview of the first shared task on “Multitask Meme Classification - Unraveling Misogynistic and Trolls in Online Memes,” organized as part of the LT-EDI@EACL 2024 conference. The task was set to classify misogynistic content and troll memes within online platforms, focusing specifically on memes in Tamil and Malayalam languages. A total of 52 teams registered for the competition, with four submitting systems for the Tamil meme classification task and three for the Malayalam task. The outcomes of this shared task are significant, providing insights into the current state of misogynistic content in digital memes and highlighting the effectiveness of various computational approaches in identifying such detrimental content. The top-performing model got a macro F1 score of 0.73 in Tamil and 0.87 in Malayalam.

Sentiment Analysis (SA) in Dravidian codemixed text is a hot research area right now. In this regard, the “Second Shared Task on SA in Code-mixed Tamil and Tulu” at Dravidian- LangTech (EACL-2024) is organized. Two tasks namely SA in Tamil-English and Tulu- English code-mixed data, make up this shared assignment. In total, 64 teams registered for the shared task, out of which 19 and 17 systems were received for Tamil and Tulu, respectively. The performance of the systems submitted by the participants was evaluated based on the macro F1-score. The best method obtained macro F1-scores of 0.260 and 0.584 for code-mixed Tamil and Tulu texts, respectively.

We participated in the Sentiment Analysis in Tamil and Tulu - DravidianLangTech 2023-RANLP 2023 task in the team name of VEL. This research focuses on addressing the challenge of detecting sentiment analysis in social media code-mixed comments written in Tamil and Tulu languages. Code-mixed text in social media often deviates from strict grammar rules and incorporates non-native scripts, making sentiment identification a complex task. To tackle this issue, we employ pre-processing techniques to remove unnecessary content and develop a model specifically designed for sentiment analysis detection. Additionally, we explore the effectiveness of traditional machine-learning models combined with feature extraction techniques. Our best model logistic regression configurations achieve impressive macro F1 scores of 0.43 on the Tamil test set and 0.51 on the Tulu test set, indicating promising results in accurately detecting instances of sentiment in code-mixed comments.