![[2509.18880] Diversity Boosts AI-Generated Text Detection](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2509.18880] Diversity Boosts AI-Generated Text Detection
Summary
The paper presents DivEye, a novel framework for detecting AI-generated text by analyzing unpredictability in text structure and vocabulary, outperforming existing methods significantly.
Why It Matters
As AI-generated text becomes more prevalent, effective detection methods are crucial to prevent misinformation and misuse in various sectors. DivEye's approach enhances interpretability and robustness, addressing limitations of current detection systems.
Key Takeaways
- DivEye improves AI-generated text detection by analyzing lexical and structural unpredictability.
- The framework outperforms existing zero-shot detectors by up to 33.2%.
- DivEye offers insights into why texts are flagged, enhancing interpretability.
- It is robust against paraphrasing and adversarial attacks.
- The method improves existing detectors' performance by up to 18.7% when used as an auxiliary signal.
Computer Science > Computation and Language arXiv:2509.18880 (cs) [Submitted on 23 Sep 2025 (v1), last revised 25 Feb 2026 (this version, v3)] Title:Diversity Boosts AI-Generated Text Detection Authors:Advik Raj Basani, Pin-Yu Chen View a PDF of the paper titled Diversity Boosts AI-Generated Text Detection, by Advik Raj Basani and 1 other authors View PDF HTML (experimental) Abstract:Detecting AI-generated text is an increasing necessity to combat misuse of LLMs in education, business compliance, journalism, and social media, where synthetic fluency can mask misinformation or deception. While prior detectors often rely on token-level likelihoods or opaque black-box classifiers, these approaches struggle against high-quality generations and offer little interpretability. In this work, we propose DivEye, a novel detection framework that captures how unpredictability fluctuates across a text using surprisal-based features. Motivated by the observation that human-authored text exhibits richer variability in lexical and structural unpredictability than LLM outputs, DivEye captures this signal through a set of interpretable statistical features. Our method outperforms existing zero-shot detectors by up to 33.2% and achieves competitive performance with fine-tuned baselines across multiple benchmarks. DivEye is robust to paraphrasing and adversarial attacks, generalizes well across domains and models, and improves the performance of existing detectors by up to 18.7% when used as ...
