semeval2024-boundary-detection

Solution for SemEval204-Task8-subtaskC. Our solution recieves best MAE score in accoradance with the leaderboard.

https://github.com/natriistorm/semeval2024-boundary-detection

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

✓
CITATION.cff file
Found CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
○
DOI references
○
Academic publication links
○
Academic email domains
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (10.0%) to scientific vocabulary

Last synced: 6 months ago · JSON representation ·

Repository

Solution for SemEval204-Task8-subtaskC. Our solution recieves best MAE score in accoradance with the leaderboard.

Basic Info

Host: GitHub
Owner: natriistorm
License: mit
Language: Python
Default Branch: main
Homepage: https://aclanthology.org/2024.semeval-1.257/
Size: 1.67 MB

Statistics

Stars: 0
Watchers: 1
Forks: 0
Open Issues: 0
Releases: 0

Created about 2 years ago · Last pushed over 1 year ago

Metadata Files

Readme License Citation

DeepPavlov at SemEval-2024 Task 8: Leveraging Transfer Learning for Detecting Boundaries of Machine-Generated Texts

[Anastasia Voznyuk](https://github.com/natriistorm)^{1 :email: *}, Vasily Konovalov¹ ¹ Moscow Institute of Physics and Technology ^:email: Corresponding author: vozniuk.ae@phystech.edu [📝 Paper](https://aclanthology.org/2024.semeval-1.257/), [ Code](https://github.com/natriistorm/SemEval2024-boundary-detection/tree/main/src)

💡 Abstract

The Multigenerator, Multidomain, and Multilingual Black-Box Machine-Generated Text Detection shared task in the SemEval-2024 competition aims to tackle the problem of misusing collaborative human-AI writing. Although there are a lot of existing detectors of AI content, they are often designed to give a binary answer and thus may not be suitable for more nuanced problem of finding the boundaries between human-written and machine-generated texts, while hybrid human-AI writing becomes more and more popular. In this paper, we address the boundary detection problem. Particularly, we present a pipeline for augmenting data for supervised fine-tuning of DeBERTaV3. We receive new best MAE score, according to the leaderboard of the competition, with this pipeline.

🔎 Overview

🛠️ Repository Structure

The repository is structured as follows: - src: This directory contains the code used in the paper and for submission. shell Forecasting-fMRI-Images ├── LICENSE ├── README.md └── code ├── run.sh # shell script to load transformer_baseline and start experiment ├── data_augmentation.py # main file for augmentation ├── transformer_baseline.py # file to run experiments ├── splitter.py # util file for splitting the texts └── scorer.py # file to calculate MAE

🔎 Citation

@inproceedings{voznyuk-konovalov-2024-deeppavlov, title = "{D}eep{P}avlov at {S}em{E}val-2024 Task 8: Leveraging Transfer Learning for Detecting Boundaries of Machine-Generated Texts", author = "Voznyuk, Anastasia and Konovalov, Vasily", booktitle = "Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024)", month = jun, year = "2024", address = "Mexico City, Mexico", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2024.semeval-1.257", pages = "1821--1829" }

Owner

Name: Anastasia Voznyuk
Login: natriistorm
Kind: user

Repositories: 2
Profile: https://github.com/natriistorm

Citation (CITATION.cff)

title: "DeepPavlov at SemEval-2024 Task 8: Leveraging Transfer Learning for
  Detecting Boundaries of Machine-Generated Texts"
abstract: The Multigenerator, Multidomain, and Multilingual Black-Box
  Machine-Generated Text Detection shared task in the SemEval-2024 competition
  aims to tackle the problem of misusing collaborative human-AI writing.
  Although there are a lot of existing detectors of AI content, they are often
  designed to give a binary answer and thus may not be suitable for more nuanced
  problem of finding the boundaries between human-written and machine-generated
  texts, while hybrid human-AI writing becomes more and more popular. In this
  paper, we address the boundary detection problem. Particularly, we present a
  pipeline for augmenting data for supervised fine-tuning of DeBERTaV3. We
  receive new best MAE score, according to the leaderboard of the competition,
  with this pipeline.
authors:
  - family-names: Voznyuk
    given-names: Anastasia
  - family-names: Konovalov
    given-names: Vasily
cff-version: 1.2.0
date-released: 2024-06-28
identifiers:
  - type: url
    value: "https://aclanthology.org/2024.semeval-1.257"
    description: Latest version
license: Apache-2.0
repository-code: https://github.com/natriistorm/SemEval2024-boundary-detection
preferred-citation:
  title: "DeepPavlov at SemEval-2024 Task 8: Leveraging Transfer Learning for
    Detecting Boundaries of Machine-Generated Texts"
  type: conference-paper
  authors:
    - family-names: Voznyuk
      given-names: Anastasia
    - family-names: Konovalov
      given-names: Vasily
  collection-title: Proceedings of the 18th International Workshop on Semantic
    Evaluation (SemEval-2024)
  collection-type: proceedings
  conference:
    name: SemEval
  start: 1821
  end: 1829
  year: 2024

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Open Source Science