Misogyny Meme Detection - DravidianLangTech@NAACL 2025

This repository contains our implementation for the Shared Task on Misogyny Meme Detection organized as part of DravidianLangTech@NAACL 2025. This task focuses on detecting misogyny in memes using multimodal machine learning models, leveraging both textual and visual elements.

Task Overview

The shared task challenges participants to classify memes as Misogynistic or Non-misogynistic in the Tamil and Malayalam languages. The datasets provided include both text (transcriptions) and image components, requiring participants to develop systems that analyze and integrate multimodal information effectively.

Key Details:

• Languages: Tamil and Malayalam
• Evaluation Metric: Macro F1 Score
• Dataset: Social media memes containing textual transcriptions and images
• Objective: Build a system to classify memes into two categories:
  • Misogynistic
  • Non-misogynistic

For more details, visit here.

Repository Structure

├── data/
│   ├── train/                # Training data (images and CSV)
│   ├── valid/                # Development data (images and CSV)
│   ├── test/                 # Test data (images and CSV)
├── notebooks/
│   ├── Tamil Submission Notebook
│   ├── Malayalam Submission Notebook
├── requirements.txt          # Required Python libraries
├── README.md                 # Project documentation

Methodology

This section will be updated very soon.

Installation

1. Clone the repository:

   git clone https://github.com/RJ-Hossan/MMD-NAACL-2025
   cd MMD-NAACL-2025

2. Install dependencies:

   pip install -r requirements.txt

3. Download the datasets from this link and place them in the data/ directory.

Acknowledgments

We thank the organizers of DravidianLangTech@NAACL 2025 for providing the datasets and hosting this important shared task. We also acknowledge the support of the open-source community for tools like PyTorch, Hugging Face Transformers, and torchvision.

---