Variation is the Key: A Variation-Based Framework for LLM-Generated Text Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915797754970112 |
|---|---|
| author | Li, Xuecong Li, Xiaohong Hu, Qiang Zhang, Yao Wang, Junjie |
| author_facet | Li, Xuecong Li, Xiaohong Hu, Qiang Zhang, Yao Wang, Junjie |
| contents | Detecting text generated by large language models (LLMs) is crucial but challenging. Existing detectors depend on impractical assumptions, such as white-box settings, or solely rely on text-level features, leading to imprecise detection ability. In this paper, we propose a simple but effective and practical LLM-generated text detection method, VaryBalance. The core of VaryBalance is that, compared to LLM-generated texts, there is a greater difference between human texts and their rewritten version via LLMs. Leveraging this observation, VaryBalance quantifies this through mean standard deviation and distinguishes human texts and LLM-generated texts. Comprehensive experiments demonstrated that VaryBalance outperforms the state-of-the-art detectors, i.e., Binoculars, by up to 34.3\% in terms of AUROC, and maintains robustness against multiple generating models and languages. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_13226 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Variation is the Key: A Variation-Based Framework for LLM-Generated Text Detection Li, Xuecong Li, Xiaohong Hu, Qiang Zhang, Yao Wang, Junjie Artificial Intelligence Computation and Language Detecting text generated by large language models (LLMs) is crucial but challenging. Existing detectors depend on impractical assumptions, such as white-box settings, or solely rely on text-level features, leading to imprecise detection ability. In this paper, we propose a simple but effective and practical LLM-generated text detection method, VaryBalance. The core of VaryBalance is that, compared to LLM-generated texts, there is a greater difference between human texts and their rewritten version via LLMs. Leveraging this observation, VaryBalance quantifies this through mean standard deviation and distinguishes human texts and LLM-generated texts. Comprehensive experiments demonstrated that VaryBalance outperforms the state-of-the-art detectors, i.e., Binoculars, by up to 34.3\% in terms of AUROC, and maintains robustness against multiple generating models and languages. |
| title | Variation is the Key: A Variation-Based Framework for LLM-Generated Text Detection |
| topic | Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2602.13226 |