Zero-Shot Detection of LLM-Generated Text via Implicit Reward Model

Runheng Liu, Heyan Huang, Xingchen Xiao, Zhijing Wu

Keywords: Large Language Model, LLM Generated Text Detection, Reward Model, Zero-Shot Detection

TL;DR: We propose an effective zero-shot method for detecting LLM-generated text by leveraging implicit reward models.

Abstract:

Large language models (LLMs) have demonstrated remarkable capabilities across various tasks. However, their ability to generate human-like text has raised concerns about potential misuse. This underscores the need for reliable and effective methods to detect LLM-generated text.
In this paper, we propose IRM, a novel zero-shot approach that leverages Implicit Reward Models for LLM-generated text detection. Such implicit reward models can be derived from publicly available instruction-tuned and base models. Previous reward-based method relies on preference construction and task-specific fine-tuning. In comparison, IRM requires neither preference collection nor additional training.
We evaluate IRM on the DetectRL benchmark and demonstrate that IRM can achieve superior detection performance, outperforms existing zero-shot and supervised methods in LLM-generated text detection.

Primary Area: Social and economic aspects of machine learning (e.g., fairness, interpretability, human-AI interaction, privacy, safety, strategic behavior)

Submission Number: 12022