Unmask It! AI-Generated Product Review Detection in Dravidian Languages

Somsubhra De, Advait Vats
3/12/2025
cs.CLcs.AIcs.LG

No official code found. Best implementation: somsubhra04/dravlangtech_ai-gen-prod-rev (0 stars, GPL-3.0, last commit 2025-03-13).

Abstract

The rise of Generative AI has led to a surge in AI-generated reviews, often posing a serious threat to the credibility of online platforms. Reviews serve as the primary source of information about products and services. Authentic reviews play a vital role in consumer decision-making. The presence of fabricated content misleads consumers, undermines trust and facilitates potential fraud in digital marketplaces. This study focuses on detecting AI-generated product reviews in Tamil and Malayalam, two low-resource languages where research in this domain is relatively under-explored. We worked on a range of approaches - from traditional machine learning methods to advanced transformer-based models such as Indic-BERT, IndicSBERT, MuRIL, XLM-RoBERTa and MalayalamBERT. Our findings highlight the effectiveness of leveraging the state-of-the-art transformers in accurately identifying AI-generated content, demonstrating the potential in enhancing the detection of fake reviews in low-resource language settings.

View on arXivView PDF

Code Implementations(1)

somsubhra04/dravlangtech_ai-gen-prod-revCommunity100%
GPL-3.0

Contains the code files for the shared task of DravidianLangTech @ NAACL2025

01Jupyter NotebookJan 21, 20251 years agoGPL-3.0

Similar Papers

Bengali Fake Review Detection using Semi-supervised Generative Adversarial Networks

85% match2023

Enhanced Review Detection and Recognition: A Platform-Agnostic Approach with Application to Online Commerce

83% match2024

CoCoNUTS: Concentrating on Content while Neglecting Uninformative Textual Styles for AI-Generated Peer Review Detection

83% match2025

Bengali Fake Reviews: A Benchmark Dataset and Detection System

83% match2023

Counter Turing Test ($CT^2$): Investigating AI-Generated Text Detection for Hindi -- Ranking LLMs based on Hindi AI Detectability Index ($ADI_{hi}$)

82% match1 repo2024

Large Language Models as 'Hidden Persuaders': Fake Product Reviews are Indistinguishable to Humans and Machines

81% match2025

Cite this paper

@article{de2025unmask,
  title  = {Unmask It! AI-Generated Product Review Detection in Dravidian Languages},
  author = {Somsubhra De and Advait Vats},
  year   = {2025},
  eprint = {2503.09289},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2503.09289v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2503.09289.svg)](https://codeofpaper.com/paper/2503.09289)

Discussion