用大规模数据训练新方法，精准识别大模型生成文本
On the Effectiveness of LLM-Specific Fine-Tuning for Detecting AI-Generated Text
- 按不同大模型或家族分别微调检测器，提升针对性
- 在覆盖21个大模型的百万级测试集上达99.6%准确率
- 适合教育、出版等领域需要验证文本真实性的场景
大语言模型的快速发展使生成文本接近人类写作水平，给教育、出版和数字安全领域的文本真实性验证带来挑战。本文基于大规模语料库与创新训练策略，开展人工智能生成文本检测的全面研究。构建了包含10亿词元的人类写作语料库（跨多种文体）和19亿词元的AI生成语料库（涵盖多种大模型与领域）。利用这些资源，开发并评估多个检测模型，提出两种新型微调范式：按模型（Per LLM）与按模型家族（Per LLM family）微调。在覆盖21个大模型的1亿词元基准测试中，最优微调检测器达到高达99.6%的词级别准确率，显著优于现有开源基线。
原文摘要 · Abstract (English)
The rapid progress of large language models has enabled the generation of text that closely resembles human writing, creating challenges for authenticity verification in education, publishing, and digital security. Detecting AI-generated text has therefore become a crucial technical and ethical issue. This paper presents a comprehensive study of AI-generated text detection based on large-scale corpora and novel training strategies. We introduce a 1-billion-token corpus of human-authored texts spanning multiple genres and a 1.9-billion-token corpus of AI-generated texts produced by prompting a variety of LLMs across diverse domains. Using these resources, we develop and evaluate numerous detection models and propose two novel training paradigms: Per LLM and Per LLM family fine-tuning. Across a 100-million-token benchmark covering 21 large language models, our best fine-tuned detector achieves up to $99.6\%$ token-level accuracy, substantially outperforming existing open-source baselines.
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。