Awesome
Paper Link
Please find our paper at https://aclanthology.org/2024.findings-acl.663/
Installation
- Clone the repository:
git clone https://github.com/justaguyalways/ToxVidLLM_ACL_2024.git
cd ToxVidLLM_ACL_2024
- Create a conda environment and activate it:
conda create --name your-env-name python=3.8
conda activate your-env-name
- Install the required packages:
pip install -r requirements.txt
Dataset
-
Download the dataset from the following link: ToxCMM Dataset Link
-
Unzip the downloaded file:
unzip dataset.zip
- Move the unzipped folder to the
final_data
directory within the repository:
mv path_to_unzipped_folder final_data
Usage
Training
To train the model, run train.py
. You can specify which GPU to use with the CUDA_VISIBLE_DEVICES
environment variable. Replace xxxx
with the appropriate GPU ID (e.g., 0
for the first GPU).
CUDA_VISIBLE_DEVICES=xxxx python train.py
Example:
CUDA_VISIBLE_DEVICES=0 python train.py
Testing
To test the model, run test.py
. Similarly, you can specify the GPU with CUDA_VISIBLE_DEVICES
.
CUDA_VISIBLE_DEVICES=xxxx python test.py
Example:
CUDA_VISIBLE_DEVICES=0 python test.py
Citation
If you use our work or find it useful, please cite:
@inproceedings{maity-etal-2024-toxvidlm,
title = "{T}ox{V}id{LM}: A Multimodal Framework for Toxicity Detection in Code-Mixed Videos",
author = "Maity, Krishanu and
Sangeetha, Poornash and
Saha, Sriparna and
Bhattacharyya, Pushpak",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand and virtual meeting",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.663",
pages = "11130--11142",
}