https://github.com/andythefactory/evalita2023
https://github.com/andythefactory/evalita2023
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/andythefactory/evalita2023
- Owner: AndyTheFactory
- License: apache-2.0
- Created: 2023-03-25T09:43:29.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2023-06-18T13:11:27.000Z (almost 2 years ago)
- Last Synced: 2024-12-27T18:27:00.226Z (5 months ago)
- Language: Jupyter Notebook
- Size: 1.11 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Evalita 2023 / ACTI Task - Detecting Conspiracy Theories in Italian
## UPB @ ACTI: Detecting Conspiracies using fine tuned Sentence Transformers
We focused on employing the power of pretrained Italian language sentence Transformers. To further enhance the
performance and address potential biases, we employed Large Language Models (LLMs) to augment the training data, resulting in a more balanced
and comprehensive training set. This combination of leveraging pre-trained models and data augmentation techniques formed the foundation of our
methodology, enabling us to achieve first place in the final leaderboard of both sub-tasks with F1 scores of 85.71% and respectively 91.23%.
```
Citation:
@inproceedings{acti2023upb,
title={UPB at ACTI: Detecting Conspiracies using fine tuned Sentence Transformers},
author={Paraschiv, Andrei and Dascalu, Mihai},
booktitle={Proceedings of the 9th Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (EVALITA 2023)},
year={2023}
}
```