Saved in:
Bibliographic Details
Main Authors: West, Alva, Weng, Yixuan, Zhu, Minjun, Zhang, Luodan, Lin, Zhen, Bao, Guangsheng, Zhang, Yue
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.01754
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909801414393856
author West, Alva
Weng, Yixuan
Zhu, Minjun
Zhang, Luodan
Lin, Zhen
Bao, Guangsheng
Zhang, Yue
author_facet West, Alva
Weng, Yixuan
Zhu, Minjun
Zhang, Luodan
Lin, Zhen
Bao, Guangsheng
Zhang, Yue
contents The field of AI-generated text detection has evolved from supervised classification to zero-shot statistical analysis. However, current approaches share a fundamental limitation: they aggregate token-level measurements into scalar scores, discarding positional information about where anomalies occur. Our empirical analysis reveals that AI-generated text exhibits significant non-stationarity, statistical properties vary by 73.8\% more between text segments compared to human writing. This discovery explains why existing detectors fail against localized adversarial perturbations that exploit this overlooked characteristic. We introduce Temporal Discrepancy Tomography (TDT), a novel detection paradigm that preserves positional information by reformulating detection as a signal processing task. TDT treats token-level discrepancies as a time-series signal and applies Continuous Wavelet Transform to generate a two-dimensional time-scale representation, capturing both the location and linguistic scale of statistical anomalies. On the RAID benchmark, TDT achieves 0.855 AUROC (7.1\% improvement over the best baseline). More importantly, TDT demonstrates robust performance on adversarial tasks, with 14.1\% AUROC improvement on HART Level 2 paraphrasing attacks. Despite its sophisticated analysis, TDT maintains practical efficiency with only 13\% computational overhead. Our work establishes non-stationarity as a fundamental characteristic of AI-generated text and demonstrates that preserving temporal dynamics is essential for robust detection.
format Preprint
id arxiv_https___arxiv_org_abs_2508_01754
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AI-Generated Text is Non-Stationary: Detection via Temporal Tomography
West, Alva
Weng, Yixuan
Zhu, Minjun
Zhang, Luodan
Lin, Zhen
Bao, Guangsheng
Zhang, Yue
Computation and Language
The field of AI-generated text detection has evolved from supervised classification to zero-shot statistical analysis. However, current approaches share a fundamental limitation: they aggregate token-level measurements into scalar scores, discarding positional information about where anomalies occur. Our empirical analysis reveals that AI-generated text exhibits significant non-stationarity, statistical properties vary by 73.8\% more between text segments compared to human writing. This discovery explains why existing detectors fail against localized adversarial perturbations that exploit this overlooked characteristic. We introduce Temporal Discrepancy Tomography (TDT), a novel detection paradigm that preserves positional information by reformulating detection as a signal processing task. TDT treats token-level discrepancies as a time-series signal and applies Continuous Wavelet Transform to generate a two-dimensional time-scale representation, capturing both the location and linguistic scale of statistical anomalies. On the RAID benchmark, TDT achieves 0.855 AUROC (7.1\% improvement over the best baseline). More importantly, TDT demonstrates robust performance on adversarial tasks, with 14.1\% AUROC improvement on HART Level 2 paraphrasing attacks. Despite its sophisticated analysis, TDT maintains practical efficiency with only 13\% computational overhead. Our work establishes non-stationarity as a fundamental characteristic of AI-generated text and demonstrates that preserving temporal dynamics is essential for robust detection.
title AI-Generated Text is Non-Stationary: Detection via Temporal Tomography
topic Computation and Language
url https://arxiv.org/abs/2508.01754