Popular GitHub Repositories for Hateful Memes

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)

Accelerate Neural Net Training by Progressively Freezing Layers

?Vilio: State-of-the-art VL models in PyTorch & PaddlePaddle

Resources (conference/journal publications, references to dataset) for harmful memes detection.

A chaotic, Python-style scripting language for .NET — powered by memes and sloths. Built with the OOPSIE framework.

Deep Learning model to predict the template of the given meme

DeepHumor: Image-based Meme Generation using Deep Learning

Machine learning model trained for classification of memes and photos from images

Easily run glif.app AI workflows inside your LLM: image generators, memes, selfies, and more. Glif supports all major multimedia AI models inside one app

A simple Model Context Protocol (MCP) server for generating memes using the ImgFlip API

? Add text overlays to segmented objects in your images using AI. Powered by Meta's SAM2 for segmentation, running entirely in your browser. Perfect for creating memes, social media content, and creative image editing. No backend required!

Bimodal and Unimodal Sentiment Analysis of Internet Memes (Image+Text)

[COLING 2025?] Evolver: Chain-of-Evolution Prompting to Boost Large Multimodal Models for Hateful Meme Detection

Roast your Twitter profile with AI-generated Severance wellness memes.

Automatically create YouTube mashups. Given videos and a text, AutoMash will cut the videos together so the speakers in the video appears to says the given text.

Models to detect hateful comments served with flask trained on Kaggle's Toxic Comment Classification Challenge dataset.

Code for all three tasks of SemEval-2021 Task 6

Hateful Memes dataset contains real hate speech. The Real Hateful Memes dataset consists of more than 10,000 newly created examples by Facebook AI.

??? Botyan is a cross-platform Telegram and Discord bot powered by a language model and featuring the persona of Kate Artivian. Botyan can chat, generate memes-demotivators, run mini-games, apply video effects, and send auto-messages. Easy to use and customizable.

:black_joker: Using multiple types of annotation extracted from hateful-memes dataset and feed those data into multi-modal transformers to achieve high accuracy.