Detecting Harmful Memes with Decoupled Understanding and Guided CoT Reasoning
-
Updated
Jul 4, 2025 - Jupyter Notebook
Detecting Harmful Memes with Decoupled Understanding and Guided CoT Reasoning
A multimodal deep learning project for classifying mental health-related memes, combining both textual and visual features.
Ordantis @ EXIST 2026 Task 2 — GEMF: Gemini-enriched multimodal fusion for sexism characterization in memes (CLEF 2026).
Analyze and search images locally using AI-powered descriptions and semantic embeddings.
This AI-powered Streamlit application, built with Google GenerativeAI, empowers you to classify potentially offensive content in images, engage in conversations with a chatbot, access random jokes and a meme stream, and explore summaries of relevant research papers, all within a user-friendly web interface.
Detect sexism in memes using Gemini-Enriched Multimodal Fusion. Rank-leading system for the EXIST 2026 task on sexism characterization.
Add a description, image, and links to the meme-classification topic page so that developers can more easily learn about it.
To associate your repository with the meme-classification topic, visit your repo's landing page and select "manage topics."