Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.15514 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908837596889088 |
|---|---|
| author | Ahmed, Sara Hammond, Tracy |
| author_facet | Ahmed, Sara Hammond, Tracy |
| contents | As large language models (LLMs) become increasingly prevalent, reliable methods for detecting AI-generated text are critical for mitigating potential risks. We introduce DependencyAI, a simple and interpretable approach for detecting AI-generated text using only the labels of linguistic dependency relations. Our method achieves competitive performance across monolingual, multi-generator, and multilingual settings. To increase interpretability, we analyze feature importance to reveal syntactic structures that distinguish AI-generated from human-written text. We also observe a systematic overprediction of certain models on unseen domains, suggesting that generator-specific writing styles may affect cross-domain generalization. Overall, our results demonstrate that dependency relations alone provide a robust signal for AI-generated text detection, establishing DependencyAI as a strong linguistically grounded, interpretable, and non-neural network baseline. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_15514 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DependencyAI: Detecting AI Generated Text through Dependency Parsing Ahmed, Sara Hammond, Tracy Computation and Language As large language models (LLMs) become increasingly prevalent, reliable methods for detecting AI-generated text are critical for mitigating potential risks. We introduce DependencyAI, a simple and interpretable approach for detecting AI-generated text using only the labels of linguistic dependency relations. Our method achieves competitive performance across monolingual, multi-generator, and multilingual settings. To increase interpretability, we analyze feature importance to reveal syntactic structures that distinguish AI-generated from human-written text. We also observe a systematic overprediction of certain models on unseen domains, suggesting that generator-specific writing styles may affect cross-domain generalization. Overall, our results demonstrate that dependency relations alone provide a robust signal for AI-generated text detection, establishing DependencyAI as a strong linguistically grounded, interpretable, and non-neural network baseline. |
| title | DependencyAI: Detecting AI Generated Text through Dependency Parsing |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.15514 |