Weak Links in LinkedIn: Enhancing Fake Profile Detection in the Age of LLMs

2025-07-21

AI Summary

Large language models (LLMs) have drastically lowered the barrier to generating realistic fake LinkedIn profiles, severely undermining the robustness of existing text-based detectors. This paper presents the first systematic evaluation of mainstream detectors under LLM-generated samples, revealing critical failure modes—specifically, false acceptance rates (FAR) as high as 42–52%. To address this, we propose a GPT-assisted adversarial training framework: it leverages GPT to synthesize high-quality adversarial examples and jointly models numerical features with multimodal text embeddings (e.g., BERT and Sentence-BERT). Ablation studies validate the efficacy of our feature fusion strategy. Experiments demonstrate that our approach reduces FAR to 1–7% while maintaining low false rejection rates (0.5–2%), significantly improving generalization and adversarial robustness. The framework establishes a scalable, interpretable paradigm for fake identity detection in the LLM era.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Large Language Models (LLMs) have made it easier to create realistic fake profiles on platforms like LinkedIn. This poses a significant risk for text-based fake profile detectors. In this study, we evaluate the robustness of existing detectors against LLM-generated profiles. While highly effective in detecting manually created fake profiles (False Accept Rate: 6-7%), the existing detectors fail to identify GPT-generated profiles (False Accept Rate: 42-52%). We propose GPT-assisted adversarial training as a countermeasure, restoring the False Accept Rate to between 1-7% without impacting the False Reject Rates (0.5-2%). Ablation studies revealed that detectors trained on combined numerical and textual embeddings exhibit the highest robustness, followed by those using numerical-only embeddings, and lastly those using textual-only embeddings. Complementary analysis on the ability of prompt-based GPT-4Turbo and human evaluators affirms the need for robust automated detectors such as the one proposed in this study.

Problem

• Detecting LLM-generated fake LinkedIn profiles effectively
• Improving robustness of text-based fake profile detectors
• Reducing false accept rates for GPT-generated profiles

Innovation

• GPT-assisted adversarial training for detection
• Combined numerical and textual embeddings
• Robust automated fake profile detectors