Skip to content
Open access

Towards Fine-grained and Robust Detection of Human–AI Co-authorship

Nov 2026 · Conference on Empirical Methods in Natural Language Processing · pp. 28981-28998 · 0 citations · 39 references
Computer Science

TL;DR

This framework reveals the mechanisms of how adversarial paraphrasing blurs the decision boundaries for co-authorship detectors, and highlights the value of discrete-class formulation and principled data augmentation for building reliable detectors of human-AI co-authorship.

Abstract

The widespread adoption of large language models (LLMs) is reshaping how textual content is produced, complicating text provenance, and raising concerns for misinformation, plagiarism, and intellectual property. While prior work on AI-generated text detection has progressed, one-dimensional human-AI co-authorship scores lack signals directly relevant to ``AI involvement''. It remains unclear how adversarial paraphrasing distorts detection signals. In this work, we formulate a discrete-class text provenance detection framework. This framework allows an abstraction of realistic co-authorship scenarios into distinct degrees of AI involvement, differentiating the AI polishing on the surface level from the AI updating the discourse structures. This framework reveals the mechanisms of how adversarial paraphrasing blurs the decision boundaries for co-authorship detectors. To improve robustness of the detectors, we augment training data with adversarial paraphrases generated by the state-of-the-art paraphrasing attack model, exposing detectors to challenging meaning-preserving rewrites. Extensive experiments across multiple datasets and generation models demonstrate that adversarial paraphrasing substantially degrades detectors trained only on benign data, while adversarial data augmentation leads to consistent robustness gains and good cross-generator transferability. Overall, our findings highlight the value of discrete-class formulation and principled data augmentation for building reliable detectors of human-AI co-authorship.

Read PDF

Similar papers

2026

eGrantha.ai at PAN 2026: Contrastive Representation Learning for Robust Generative AI Authorship Detection

Ebed-KNN is presented, a system that fine-tunes RoBERTa-base with a supervised contrastive objective to produce geometrically separable embeddings of human and AI-generated text, indicating that contrastively learned embedding spaces offer a competitive and robust foundation for AI-generated text detection.

Anish Thapaliya, Shushanta Pudasaini, B. Bal · 1 citation
Preprint Nov 2026

Specificity Bias and the Wikipedia Paradox: An xAI Analysis of AI Text Detectors under Domain Shift

The widespread adoption of Large Language Models (LLMs) demands reliable detection of human-authored versus synthetic text. However, current Transformer-based detectors often experience severe performance degradation during domain shifts and operate as opaque classifiers. This paper investigates the robustness and inte...

Nico Morloher, Johannes Reschke, Mohamed Mady · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.