arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.28471cs.CLcs.LG

对比学习用于作者身份验证

Contrastive Learning for Authorship Verification

  • Georgia Institute of Technology(佐治亚理工学院)

机构由 AI 辅助整理,请以论文原文为准。

Peter Kirby

中文总结 AI 辅助

本文提出用对比学习改进作者身份验证,识别关键影响因素,并开发ModernBERT双编码器模型,在PAN21任务上达到98.4%准确率。

中文摘要 AI 辅助

我们的结果表明,在测试设置下,对比学习在作者身份验证任务中优于基于分类的方法。我们确定了损失函数、批大小、训练时长、预训练模型、输入上下文长度和随机文本跨度数据增强是影响模型性能的重要因素。基于这些考虑,我们开发了一个ModernBERT双编码器模型,在PAN21作者身份验证任务上达到了98.4%的准确率。

英文摘要

Our results show that contrastive learning outperforms a classification-based approach to authorship verification under the tested settings. We identify loss function, batch size, training duration, pre-trained model, input context length, and random text span data augmentation as important factors of model performance. Based on these considerations, we develop a ModernBERT Bi-Encoder model that achieves 98.4% accuracy on the PAN21 authorship verification task.

补充信息

↑