Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Jiatao, Ye, Mao, Peng, Cheng, Yin, Xunjian, Wan, Xiaojun
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2505.15261
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908373041020928
author Li, Jiatao
Ye, Mao
Peng, Cheng
Yin, Xunjian
Wan, Xiaojun
author_facet Li, Jiatao
Ye, Mao
Peng, Cheng
Yin, Xunjian
Wan, Xiaojun
contents Existing AI-generated text detection methods heavily depend on large annotated datasets and external threshold tuning, restricting interpretability, adaptability, and zero-shot effectiveness. To address these limitations, we propose AGENT-X, a zero-shot multi-agent framework informed by classical rhetoric and systemic functional linguistics. Specifically, we organize detection guidelines into semantic, stylistic, and structural dimensions, each independently evaluated by specialized linguistic agents that provide explicit reasoning and robust calibrated confidence via semantic steering. A meta agent integrates these assessments through confidence-aware aggregation, enabling threshold-free, interpretable classification. Additionally, an adaptive Mixture-of-Agent router dynamically selects guidelines based on inferred textual characteristics. Experiments on diverse datasets demonstrate that AGENT-X substantially surpasses state-of-the-art supervised and zero-shot approaches in accuracy, interpretability, and generalization.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15261
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AGENT-X: Adaptive Guideline-based Expert Network for Threshold-free AI-generated teXt detection
Li, Jiatao
Ye, Mao
Peng, Cheng
Yin, Xunjian
Wan, Xiaojun
Computation and Language
Existing AI-generated text detection methods heavily depend on large annotated datasets and external threshold tuning, restricting interpretability, adaptability, and zero-shot effectiveness. To address these limitations, we propose AGENT-X, a zero-shot multi-agent framework informed by classical rhetoric and systemic functional linguistics. Specifically, we organize detection guidelines into semantic, stylistic, and structural dimensions, each independently evaluated by specialized linguistic agents that provide explicit reasoning and robust calibrated confidence via semantic steering. A meta agent integrates these assessments through confidence-aware aggregation, enabling threshold-free, interpretable classification. Additionally, an adaptive Mixture-of-Agent router dynamically selects guidelines based on inferred textual characteristics. Experiments on diverse datasets demonstrate that AGENT-X substantially surpasses state-of-the-art supervised and zero-shot approaches in accuracy, interpretability, and generalization.
title AGENT-X: Adaptive Guideline-based Expert Network for Threshold-free AI-generated teXt detection
topic Computation and Language
url https://arxiv.org/abs/2505.15261