A Distribution-Free Framework for Rewrite-Based Human-text Detection via Knockoff Filtering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Liu, Yi
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916068864294912
author Liu, Yi
author_facet Liu, Yi
contents We propose a distribution-free statistical framework that converts arbitrary rewrite-based detectors into detectors with finite-sample FDR guarantees without retraining. Our key observation is that rewrite-based detection implicitly constructs knockoff samples, enabling LLM-generated text detection to be formulated as a multiple hypothesis testing problem with knockoff structure. This perspective separates the design of detection statistics from the control of false discoveries, allowing existing rewrite detectors to inherit finite-sample false discovery rate (FDR) guarantees through a simple calibration procedure. We demonstrate reliable FDR control with meaningful detection power across three detection models, 19 domains, and four LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2606_00402
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Distribution-Free Framework for Rewrite-Based Human-text Detection via Knockoff Filtering
Liu, Yi
Methodology
Artificial Intelligence
Applications
We propose a distribution-free statistical framework that converts arbitrary rewrite-based detectors into detectors with finite-sample FDR guarantees without retraining. Our key observation is that rewrite-based detection implicitly constructs knockoff samples, enabling LLM-generated text detection to be formulated as a multiple hypothesis testing problem with knockoff structure. This perspective separates the design of detection statistics from the control of false discoveries, allowing existing rewrite detectors to inherit finite-sample false discovery rate (FDR) guarantees through a simple calibration procedure. We demonstrate reliable FDR control with meaningful detection power across three detection models, 19 domains, and four LLMs.
title A Distribution-Free Framework for Rewrite-Based Human-text Detection via Knockoff Filtering
topic Methodology
Artificial Intelligence
Applications
url https://arxiv.org/abs/2606.00402