用可解释的模型检测AI生成文本，提升透明度与可信度。
DeBERTa-Sentinel: Toward Transparent and Trustworthy Detection of AI-Generated Text
- 基于DeBERTa-v3的解耦注意力机制捕捉生成文本的细微结构异常。
- 在2.8万样本上达到98.21%准确率，误检率仅0.665%。
- 提供逐词解释，适合记者、教育者等需要审计判断的群体使用。
大规模语言模型在网上的快速传播引发虚假信息、学术诚信、自动化内容操纵及对弱势网络社区的风险。现有基于Transformer的检测器如GPT-Sentinel虽有潜力，但难以泛化至多样模型输出和改写攻击，限制了其在构建可信网络生态中的作用。本文提出DeBERTa-Sentinel，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。核心设计原则为透明性：与黑箱商业检测器不同，DeBERTa-Sentinel暴露其决策的逐词解释，使记者、教育者及平台安全团队能够审计、质疑和上下文化检测结果。基于包含28,057个样本（人类与LLM生成，涵盖GPT、LLaMA、Claude）的GLC-AIText数据集（60-20-20划分），DeBERTa-Sentinel在验证集上达98.21%准确率，超越NeurIPS 2025的RoBERTa-Sentinel基线，在测试集上实现97.53%准确率、95.89%精确率、99.33%召回率、99.53%ROC-AUC，且保持0.665%的低误检率。模型可解释性揭示了学术化措辞和正式过渡词等与合成文本相关的语言标记，直接满足利益相关方对可验证、可审计内容真实性判断的需求。通过减少偏见并增强可解释性，DeBERTa-Sentinel推动可信、伦理且以人为本的AI系统发展。代码与数据见https://github.com/Galileo-Galili/HUMAN-VS-AI-TEXT-DETECTION。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。