操作引导的渐进式人机文本转换基准：面向多粒度AI文本检测
Operation-Guided Progressive Human-to-AI Text Transformation Benchmark for Multi-Granularity AI-Text Detection
AI总结：
提出OpAI-Bench基准，通过九种渐进修订版本和五种AI编辑操作，模拟人机协作编辑过程，支持文档、句子、词元和跨度多粒度检测，揭示AI文本可检测性受编辑操作、领域和累积修订历史影响，并发现混合作者中间版本比纯人类或纯AI端点更难检测。
摘要：
随着AI写作助手越来越多地融入现实世界的起草和修订流程，许多文档不再是纯粹的人类撰写或AI生成，而是渐进式人机共同编辑的结果。然而，现有的AI文本检测基准主要关注最终输出，对AI作者身份信号如何在修订过程中出现、累积或消失的理解有限。我们引入了OpAI-Bench，一个操作引导的基准，用于研究在文档、句子、词元和跨度粒度上的渐进式人机文本转换。从人类撰写的文档开始，OpAI-Bench在预定义的AI覆盖水平和五种代表性AI编辑操作下，为每个样本构建了九个顺序修订版本，涵盖四个领域，同时保留多粒度上的完整作者身份来源。该基准支持8个文档级检测器、7个句子级检测器和2个细粒度词元/跨度级检测器的全面评估。实验表明，AI文本的可检测性不仅受AI编辑内容比例的影响，还受编辑操作、领域和累积修订历史的影响。有趣的是，我们注意到混合作者身份的中间版本通常比完全人类或大量AI编辑的端点更难检测，暴露了现有基准遗漏的非单调检测模式。OpAI-Bench为分析在现实渐进编辑场景下，AI辅助写作是否、何时以及如何变得可检测提供了一个受控测试平台。我们的代码和基准可在https://github.com/VILA-Lab/OpAI-Bench获取。
英文摘要：
As AI writing assistants become increasingly integrated into real-world drafting and revision workflows, many documents are no longer purely human-written or AI-generated, but instead result from progressive human-AI co-editing. However, existing AI-text detection benchmarks largely focus on final outputs and provide limited understanding of how AI authorship signals emerge, accumulate, or disappear throughout the revision process. We introduce OpAI-Bench, an operation-guided benchmark for studying progressive human-to-AI text transformation across document, sentence, token, and span granularities. Starting from human-written documents, OpAI-Bench constructs nine sequentially revised versions for each sample under predefined AI coverage levels and five representative AI edit operations, covering four domains while preserving complete authorship provenance at multiple granularities. The benchmark supports comprehensive evaluation with 8 document-level detectors, 7 sentence-level detectors, and 2 fine-grained token/span-level detectors. Experiments reveal that AI-text detectability is governed not only by the proportion of AI-edited content, but also by edit operation, domain, and cumulative revision history. Interestingly, we notice that mixed-authorship intermediate versions are often harder to detect than both fully human and heavily AI-edited endpoints, exposing non-monotonic detection patterns missed by existing benchmarks. OpAI-Bench provides a controlled testbed for analyzing whether, when, and how AI-assisted writing becomes detectable under realistic progressive editing scenarios. Our code and benchmark are available at https://github.com/VILA-Lab/OpAI-Bench.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
