Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.04838 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917182380703744 |
|---|---|
| author | Teja, L. D. M. S. Sai Krishna, N. Siva Gopala Khan, Ufaq Khan, Muhammad Haris Mishra, Atul |
| author_facet | Teja, L. D. M. S. Sai Krishna, N. Siva Gopala Khan, Ufaq Khan, Muhammad Haris Mishra, Atul |
| contents | In the age of advanced large language models (LLMs), the boundaries between human and AI-generated text are becoming increasingly blurred. We address the challenge of segmenting mixed-authorship text, that is identifying transition points in text where authorship shifts from human to AI or vice-versa, a problem with critical implications for authenticity, trust, and human oversight. We introduce a novel framework, called Info-Mask for mixed authorship detection that integrates stylometric cues, perplexity-driven signals, and structured boundary modeling to accurately segment collaborative human-AI content. To evaluate the robustness of our system against adversarial perturbations, we construct and release an adversarial benchmark dataset Mixed-text Adversarial setting for Segmentation (MAS), designed to probe the limits of existing detectors. Beyond segmentation accuracy, we introduce Human-Interpretable Attribution (HIA overlays that highlight how stylometric features inform boundary predictions, and we conduct a small-scale human study assessing their usefulness. Across multiple architectures, Info-Mask significantly improves span-level robustness under adversarial conditions, establishing new baselines while revealing remaining challenges. Our findings highlight both the promise and limitations of adversarially robust, interpretable mixed-authorship detection, with implications for trust and oversight in human-AI co-authorship. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_04838 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DAMASHA: Detecting AI in Mixed Adversarial Texts via Segmentation with Human-interpretable Attribution Teja, L. D. M. S. Sai Krishna, N. Siva Gopala Khan, Ufaq Khan, Muhammad Haris Mishra, Atul Computation and Language In the age of advanced large language models (LLMs), the boundaries between human and AI-generated text are becoming increasingly blurred. We address the challenge of segmenting mixed-authorship text, that is identifying transition points in text where authorship shifts from human to AI or vice-versa, a problem with critical implications for authenticity, trust, and human oversight. We introduce a novel framework, called Info-Mask for mixed authorship detection that integrates stylometric cues, perplexity-driven signals, and structured boundary modeling to accurately segment collaborative human-AI content. To evaluate the robustness of our system against adversarial perturbations, we construct and release an adversarial benchmark dataset Mixed-text Adversarial setting for Segmentation (MAS), designed to probe the limits of existing detectors. Beyond segmentation accuracy, we introduce Human-Interpretable Attribution (HIA overlays that highlight how stylometric features inform boundary predictions, and we conduct a small-scale human study assessing their usefulness. Across multiple architectures, Info-Mask significantly improves span-level robustness under adversarial conditions, establishing new baselines while revealing remaining challenges. Our findings highlight both the promise and limitations of adversarially robust, interpretable mixed-authorship detection, with implications for trust and oversight in human-AI co-authorship. |
| title | DAMASHA: Detecting AI in Mixed Adversarial Texts via Segmentation with Human-interpretable Attribution |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.04838 |