Detection of Illicit Content on Online Marketplaces using Large Language Models

TL;DR

This paper evaluates the effectiveness of large language models like Llama 3.2 and Gemma 3 in detecting illicit content on online marketplaces, demonstrating their superiority in complex, multi-category classification tasks over traditional models.

Contribution

It introduces the application of LLMs with fine-tuning techniques for illicit content detection, showing significant improvements in multi-class classification over baseline models.

Findings

LLMs perform comparably to traditional models in binary classification.

Llama 3.2 outperforms baselines in multi-class, imbalanced classification.

Fine-tuning enhances LLMs' effectiveness in illicit content detection.

Abstract

Online marketplaces, while revolutionizing global commerce, have inadvertently facilitated the proliferation of illicit activities, including drug trafficking, counterfeit sales, and cybercrimes. Traditional content moderation methods such as manual reviews and rule-based automated systems struggle with scalability, dynamic obfuscation techniques, and multilingual content. Conventional machine learning models, though effective in simpler contexts, often falter when confronting the semantic complexities and linguistic nuances characteristic of illicit marketplace communications. This research investigates the efficacy of Large Language Models (LLMs), specifically Meta's Llama 3.2 and Google's Gemma 3, in detecting and classifying illicit online marketplace content using the multilingual DUTA10K dataset. Employing fine-tuning techniques such as Parameter-Efficient Fine-Tuning (PEFT) and quantization, these models were systematically benchmarked against a foundational transformer-based model (BERT) and traditional machine learning baselines (Support Vector Machines and Naive Bayes). Experimental results reveal a task-dependent advantage for LLMs. In binary classification (illicit vs. non-illicit), Llama 3.2 demonstrated performance comparable to traditional methods. However, for complex, imbalanced multi-class classification involving 40 specific illicit categories, Llama 3.2 significantly surpassed all baseline models. These findings offer substantial practical implications for enhancing online safety, equipping law enforcement agencies, e-commerce platforms, and cybersecurity specialists with more effective, scalable, and adaptive tools for illicit content detection and moderation.