This repository contains the proposed work on HOLD from DravidianLangtech-2024@EACL HOLD-DravidianLangTech-2024:
HOLD-Project: Hate and Offensive Language Detection in Telugu Codemixed Text
Paper Link: https://aclanthology.org/2024.dravidianlangtech-1.22/
Hateful online content is a growing concern, especially for young people. While social media platforms aim to connect us, they can also become breeding grounds for negativity and harmful language. This study tackles this issue by proposing a novel framework called HOLD, specifically designed to detect hate and offensive comments in Telugu-English(Tenglish) code-mixed social media content. HOLD leverages a combination of approaches, including three powerful models: LSTM architecture, Zypher, and openchat_3.5. The study highlights the effectiveness of prompt engineering and Quantized Low-Rank Adaptation (QLoRA) in boosting performance of 7B Models.