SEFD: Semantic-Enhanced Framework for Detecting LLM-Generated Text

TL;DR

SEFD is a novel semantic-enhanced framework that improves detection of LLM-generated text, especially in paraphrasing cases, by integrating retrieval-based techniques with traditional detectors for better accuracy and robustness.

Contribution

The paper introduces a retrieval-based semantic enhancement framework for LLM detection, improving accuracy in paraphrasing scenarios over existing methods.

Findings

Significantly improves detection accuracy in paraphrasing cases

Maintains robustness across various LLM-generated texts

Effective in real-world online forum and Q&A scenarios

Abstract

The widespread adoption of large language models (LLMs) has created an urgent need for robust tools to detect LLM-generated text, especially in light of \textit{paraphrasing} techniques that often evade existing detection methods. To address this challenge, we present a novel semantic-enhanced framework for detecting LLM-generated text (SEFD) that leverages a retrieval-based mechanism to fully utilize text semantics. Our framework improves upon existing detection methods by systematically integrating retrieval-based techniques with traditional detectors, employing a carefully curated retrieval mechanism that strikes a balance between comprehensive coverage and computational efficiency. We showcase the effectiveness of our approach in sequential text scenarios common in real-world applications, such as online forums and Q\&A platforms. Through comprehensive experiments across various LLM-generated texts and detection methods, we demonstrate that our framework substantially enhances detection accuracy in paraphrasing scenarios while maintaining robustness for standard LLM-generated content.