A Systematic Analysis of Linguistic Features in AI-Generated Text Detection Across Domains and Models
For researchers and practitioners needing interpretable AI-text detection, this work clarifies which linguistic signals generalize across models and domains, reducing reliance on fragile features.
This study systematically evaluates 284 linguistic features across 27 LLMs and 10 domains, finding that lexical richness is a robust indicator of AI-generated text, while many other features are context-dependent.
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.