This repository contains the experiments and results for the paper "Exploring Multimodal Models for Humor Recognition in Portuguese" accepted to the 16th International Conference on Computational Processing of Portuguese (PROPOR 2024).
The main focus of this paper is to explore multimodal approaches in Humor Recognition by combining textual and explicit features. According to the architecture shown below.
flowchart BT
text(Input text)
transformer(Transformer)
input(Input features)
combination(Combination module)
classification(Classification head)
text-->transformer
transformer-->combination;
input--->combination;
combination-->classification;
The corpus is available on GitHub.
This project is available as a python module to make installation and re-use easier.
Using Pipenv, just run:
pipenv install
If you prefer to use pip, run:
pip install -r requirements.txt
pip install .
The repository is organized as follows:
multimodal-humor-recognition
├───results [Final prediction files and images used in the paper]
├───scripts [Jupyter notebooks of the experiments]
└───src [Multimodal models implementation]
@inproceedings{InacioGoncaloOliveira2024,
title = {Exploring Multimodal Models for Humor Recognition in {{Portuguese}}},
booktitle = {Proceedings of the 16th International Conference on Computational Processing of Portuguese},
author = {In{\'a}cio, Marcio and Gon{\c c}alo Oliveira, Hugo},
editor = {Gamallo, Pablo and Claro, Daniela and Teixeira, Ant{\'o}nio and Real, Livy and Garcia, Marcos and Oliveira, Hugo Gon{\c c}alo and Amaro, Raquel},
year = {2024},
month = mar,
pages = {568--574},
publisher = {Association for Computational Lingustics},
address = {Santiago de Compostela, Galicia/Spain},
url = {https://aclanthology.org/2024.propor-1.62},
copyright = {All rights reserved}
}