A Systematic Analysis of Linguistic Features in AI-Generated Text Detection Across Domains and Models

A large-scale empirical study analyzing 284 linguistic features across 27 large language models and ten text domains found that classifiers based solely on these features can reliably distinguish AI-generated from human-written text. However, most previously proposed indicators proved context-dependent, with only measures of lexical richness remaining robust signals across model families and text domains. The findings establish which linguistic signals generalize across contexts, providing a foundation for more reliable and interpretable detection of machine-generated language.

arXiv:2606.04177v1 Announce Type: new Abstract: Interpretable linguistic features offer a promising approach for explaining why a given text appears machine-generated, particularly for non-expert users. However, existing findings on which features reliably indicate LLM-generated text remain fragmented across feature sets, models, and text domains. To address this gap, we conduct a large-scale empirical study assessing the robustness of linguistic signals for characterizing AI-generated text. Our analysis covers 284 interpretable linguistic features across outputs from 27 LLMs and ten text domains under cross-model and cross-domain generalization settings. We show that classifiers based solely on linguistic features can reliably distinguish AI-generated from human-written text. However, many previously proposed indicators prove strongly context-dependent, with the exception of measures of lexical richness, which remain robust signals across model families and text domains. These results demonstrate which linguistic signals generalize across contexts and provide a foundation for more reliable, interpretable analyses of AI-generated language.