科学文献中扭曲短语的检测
Detection of tortured phrases in scientific literature
AI总结：
本文提出基于语言模型的自动检测方法，通过嵌入相似性或掩码标记预测识别科学文献中的扭曲短语，最佳方法在召回率0.87和精确率0.61下有效发现新短语。
摘要：
本文提出了多种自动检测方法，用于从科学论文中提取所谓的“扭曲短语”。这些扭曲短语，例如用“flag to clamor”代替“signal to noise”，是使用改写工具以逃避抄袭检测的结果。我们构建了一个数据集，并评估了多种策略来标记此前未记录的扭曲短语。所提出并测试的方法基于语言模型，并依赖于嵌入相似性或掩码标记的预测。我们发现，使用标记预测并将分数传播到块级别的方法效果最佳。其召回率为0.87，精确率为0.61，能够检索出新的扭曲短语，供领域专家提交验证。
英文摘要：
This paper presents various automatic detection methods to extract so called tortured phrases from scientific papers. These tortured phrases, e.g. flag to clamor instead of signal to noise, are the results of paraphrasing tools used to escape plagiarism detection. We built a dataset and evaluated several strategies to flag previously undocumented tortured phrases. The proposed and tested methods are based on language models and either on embeddings similarities or on predictions of masked token. We found that an approach using token prediction and that propagates the scores to the chunk level gives the best results. With a recall value of .87 and a precision value of .61, it could retrieve new tortured phrases to be submitted to domain experts for validation.
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。
