https://github.com/deezer/quote_av
Code to reproduce the experiments presented in "Distinguishing Fictional Voices: a Study of Authorship Verification Models for Quotation Attribution" (LaTeCH-CLfl workshop at EACL 2024)
https://github.com/deezer/quote_av
Last synced: 9 months ago
JSON representation
Code to reproduce the experiments presented in "Distinguishing Fictional Voices: a Study of Authorship Verification Models for Quotation Attribution" (LaTeCH-CLfl workshop at EACL 2024)
- Host: GitHub
- URL: https://github.com/deezer/quote_av
- Owner: deezer
- Created: 2024-01-29T13:31:12.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-12-29T16:20:25.000Z (about 1 year ago)
- Last Synced: 2025-02-10T18:14:52.788Z (11 months ago)
- Language: Python
- Size: 153 MB
- Stars: 0
- Watchers: 5
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Distinguishing Fictional Voices: a Study of Authorship Verification Models for Quotation Attribution
This is the official repository for the Latech-Clfl 2024 (*EACL 2024*) paper ["Distinguishing Fictional Voices: a Study of Authorship Verification Models for Quotation Attribution"](https://aclanthology.org/2024.latechclfl-1.15). It contains all code and data to reproduce our results.
## Installation
### Data
Start by downloading the [Project Dialogism Novel Corpus](https://github.com/Priya22/project-dialogism-novel-corpus):
```bash
git clone https://github.com/Priya22/project-dialogism-novel-corpus.git
```
### Python Environment
Run the following commands to create an environment and install all the required packages:
```bash
python3 -m venv quote_av
. ./quote_av/bin/activate
pip3 install -U pip
pip3 install -r requirements.txt
```
## Running Experiments
The following will run the main experiments necessary to reproduce our results.
```bash
python main.py --experiment all --data_path project-dialogism-novel-corpus/data/ --model all --result_path results/
```
You can also run experiments using any model from Huggingface with the following. Note that the model must be similar to LUAR in the way it processes data.
```bash
python main.py --experiment all --data_path project-dialogism-novel-corpus/data/ --huggingface_model path/to_hgface_model --result_path results/
```