融合多模型嵌入，精准区分人类与AI写作文本
Enhancing Authorship Attribution through Embedding Fusion: A Novel Approach with Masked and Encoder-Decoder Language Models
- 用多个语言模型嵌入融合提升文本特征表达
- 在5个主流大模型数据上准确率超96%，MCC超0.93
- 适合需要高精度文本来源识别的场景
随着AI生成内容与人工写作文本的共存日益普遍，可靠的内容来源区分方法变得愈发重要。为此，我们提出一种基于预训练语言模型（PLMs）文本嵌入的新框架，用于区分AI生成与人工撰写的文本。该方法通过嵌入融合，整合多个语言模型的语义信息，利用其互补优势以提升判别性能。在多个公开的多样化数据集上进行的广泛评估表明，所提方法表现优异，在由五个知名大语言模型（LLMs）生成的平衡数据集上，分类准确率超过96%，马修斯相关系数（MCC）超过0.93。该结果验证了本方法的有效性与鲁棒性。
原文摘要 · Abstract (English)
The increasing prevalence of AI-generated content alongside human-written text underscores the need for reliable discrimination methods. To address this challenge, we propose a novel framework with textual embeddings from Pre-trained Language Models (PLMs) to distinguish AI-generated and human-authored text. Our approach utilizes Embedding Fusion to integrate semantic information from multiple Language Models, harnessing their complementary strengths to enhance performance. Through extensive evaluation across publicly available diverse datasets, our proposed approach demonstrates strong performance, achieving classification accuracy greater than 96% and a Matthews Correlation Coefficient (MCC) greater than 0.93. This evaluation is conducted on a balanced dataset of texts generated from five well-known Large Language Models (LLMs), highlighting the effectiveness and robustness of our novel methodology.
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。