发表机构
University of Hertfordshire; James Cook University（赫特福德大学; 詹姆斯库克大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
DeBERTa-Sentinel是基于DeBERTa-v3的透明AI生成文本检测框架，在GLC-AIText数据集上实现优异检测性能，可公开token级解释以支持利益相关者审计。
AI 中文摘要
大型语言模型（LLMs）在网络上的快速传播引发了人们对错误信息、学术诚信、自动内容操纵以及弱势网络社区风险的担忧。现有的基于Transformer的检测器，如GPT-Sentinel，虽有应用前景，但难以泛化到不同模型的输出和 paraphrasing 攻击，限制了其在构建可信网络生态系统中的作用。本研究提出DeBERTa-Sentinel，这是一种负责任的AI生成文本检测框架，利用DeBERTa-v3的解耦注意力机制捕捉合成内容中的细微结构异常。其核心设计原则是透明度：与黑箱式商业检测器不同，DeBERTa-Sentinel会公开其决策的 token 级解释，使受影响的利益相关者——记者、教育工作者以及平台信任与安全团队——能够对检测结果进行审计、质疑和情境化分析。使用包含28057个人类及LLM生成样本（来自GPT、LLaMA和Claude）的GLC-AIText数据集，按60-20-20划分，DeBERTa-Sentinel的验证准确率达98.21%，超过了NeurIPS 2025的RoBERTa-Sentinel基线，测试准确率为97.53%，精确率为95.89%，召回率为99.33%，ROC-AUC为99.53%，同时保持0.665%的假阴性率。该模型的可解释性揭示了与合成文本相关的语言标记，如学术措辞和正式过渡，直接支持利益相关者对可验证、可审计的内容真实性决策的需求。通过推进减少偏差、增强可解释性的负责任检测方法，DeBERTa-Sentinel推动了可信、合乎道德且以人类为中心的AI系统发展。代码和数据可在指定URL获取。
英文摘要
The rapid spread of large language models (LLMs) across the web raises concerns about misinformation, academic integrity, automated content manipulation, and risks to vulnerable online communities. Existing transformer-based detectors, such as GPT-Sentinel, show promise but struggle to generalize to diverse model outputs and paraphrasing attacks, limiting their role in building trustworthy web ecosystems. This work introduces DeBERTa-Sentinel, a responsible AI-generated text detection framework leveraging DeBERTa-v3's disentangled attention to capture subtle structural irregularities in synthetic content. A central design principle is transparency: unlike black-box commercial detectors, DeBERTa-Sentinel exposes token-level explanations of its decisions, enabling affected stakeholders journalists, educators, and platform trust and safety teams to audit, challenge, and contextualize detection outcomes. Using the GLC-AIText dataset of 28,057 human and LLM-generated samples (GPT, LLaMA, and Claude) with a 60-20-20 split, DeBERTa-Sentinel achieves 98.21\% validation accuracy and surpasses the RoBERTa-Sentinel baseline from NeurIPS 2025, achieving 97.53\% test accuracy, 95.89\% precision, 99.33\% recall, and 99.53\% ROC-AUC, and maintaining a 0.665\% false negative rate. The model's interpretability reveals linguistic markers such as academic phrasing and formal transitions associated with synthetic text, directly supporting stakeholder needs for verifiable, auditable content-authenticity decisions. By advancing responsible detection methods that reduce bias and enhance explainability, DeBERTa-Sentinel promotes trustworthy, ethical, and human-centric AI systems. Code and data are available at https://github.com/Galileo-Galili/HUMAN-VS-AI-TEXT-DETECTION.
