UVCE-IIITT@DravidianLangTech-EACL2021: Tamil Troll Meme Classification: You need to Pay more Attention

TL;DR

This paper presents a transformer-based model for classifying Tamil memes as troll or non-troll by analyzing both images and captions, aiming to improve accuracy by focusing on relevant features and ignoring noise.

Contribution

It introduces a novel transformer-transformer architecture specifically designed for multimodal meme classification in Tamil, leveraging attention mechanisms for enhanced feature extraction.

Findings

Achieved state-of-the-art accuracy on Tamil meme dataset

Demonstrated effectiveness of attention in noise reduction

Validated model's robustness across multimodal data

Abstract

Tamil is a Dravidian language that is commonly used and spoken in the southern part of Asia. In the era of social media, memes have been a fun moment in the day-to-day life of people. Here, we try to analyze the true meaning of Tamil memes by categorizing them as troll and non-troll. We propose an ingenious model comprising of a transformer-transformer architecture that tries to attain state-of-the-art by using attention as its main component. The dataset consists of troll and non-troll images with their captions as text. The task is a binary classification task. The objective of the model is to pay more attention to the extracted features and to ignore the noise in both images and text.