Introducing a multimodal model leveraging a caption generation step and hierarchical classification to identify persuasive elements in memes, aimed at combating disinformation and propaganda on social media.

BCAmirs at SemEval-2024 Task 4: Beyond Words: A Multimodal and Multilingual Exploration of Persuasion in Memes

University of British Columbia

roberta

misinformation detection

memes

caption generation

clip

vision-language models

persuasion techniques

gpt-4

multimodal analysis

hierarchical multi-label classification

Amirhossein Abaskohi

1

SHORT BIO

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES