Saved in:
Bibliographic Details
Main Authors: Li, Yang, Sheng, Qiang, Wang, Zhengjia, Yang, Yehan, Wang, Danding, Cao, Juan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.04932
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909053677993984
author Li, Yang
Sheng, Qiang
Wang, Zhengjia
Yang, Yehan
Wang, Danding
Cao, Juan
author_facet Li, Yang
Sheng, Qiang
Wang, Zhengjia
Yang, Yehan
Wang, Danding
Cao, Juan
contents The misuse of large language models (LLMs) requires precise detection of synthetic text. Existing works mainly follow binary or ternary classification settings, which can only distinguish pure human/LLM text or collaborative text at best. This remains insufficient for the nuanced regulation, as the LLM-polished human text and humanized LLM text often trigger different policy consequences. In this paper, we explore fine-grained LLM-generated text detection under a rigorous four-class setting. To handle such complexities, we propose RACE (Rhetorical Analysis for Creator-Editor Modeling), a fine-grained detection method that characterizes the distinct signatures of creator and editor. Specifically, RACE utilizes Rhetorical Structure Theory (RST) to construct a logic graph for the creator's foundation while extracting Elementary Discourse Unit (EDU)-level features for the editor's style. Experiments show that RACE outperforms 12 baselines in identifying fine-grained types with low false alarms, offering a policy-aligned solution for LLM regulation.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04932
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Beyond the Final Actor: Modeling the Dual Roles of Creator and Editor for Fine-Grained LLM-Generated Text Detection
Li, Yang
Sheng, Qiang
Wang, Zhengjia
Yang, Yehan
Wang, Danding
Cao, Juan
Computation and Language
The misuse of large language models (LLMs) requires precise detection of synthetic text. Existing works mainly follow binary or ternary classification settings, which can only distinguish pure human/LLM text or collaborative text at best. This remains insufficient for the nuanced regulation, as the LLM-polished human text and humanized LLM text often trigger different policy consequences. In this paper, we explore fine-grained LLM-generated text detection under a rigorous four-class setting. To handle such complexities, we propose RACE (Rhetorical Analysis for Creator-Editor Modeling), a fine-grained detection method that characterizes the distinct signatures of creator and editor. Specifically, RACE utilizes Rhetorical Structure Theory (RST) to construct a logic graph for the creator's foundation while extracting Elementary Discourse Unit (EDU)-level features for the editor's style. Experiments show that RACE outperforms 12 baselines in identifying fine-grained types with low false alarms, offering a policy-aligned solution for LLM regulation.
title Beyond the Final Actor: Modeling the Dual Roles of Creator and Editor for Fine-Grained LLM-Generated Text Detection
topic Computation and Language
url https://arxiv.org/abs/2604.04932