arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.11196cs.CL
收藏

协变量偏移下作者身份验证的自校正神经概率模型

Self-Calibrating Neural-Probabilistic Model for Authorship Verification Under Covariate Shift

Ruhr University Bochum(波鸿鲁尔大学);Bucknell University(巴克内尔大学)

Benedikt Boenninghoff, Dorothea Kolossa, Robert M. Nickel

摘要:

我们正在解决两个基本问题,在作者身份验证(AV):主题的变化和误校准。两个有争议的文本的主题的变化是大多数AV系统错误的主要原因。此外,观察到由深度学习AV机制产生的潜在概率估计常常与相应训练数据中的实际案例计数不匹配。因此,概率估计的校准很差。我们正在从PAN 2020扩展我们的框架,以包括贝叶斯因子评分(BFS)和不确定性适应层(UAL)来解决这两个问题。使用2020/21 PAN AV共享任务数据的实验表明,所提出的方法显着降低了对主题变化的敏感性,并显着提高了系统的校准。

英文摘要:

We are addressing two fundamental problems in authorship verification (AV): Topic variability and miscalibration. Variations in the topic of two disputed texts are a major cause of error for most AV systems. In addition, it is observed that the underlying probability estimates produced by deep learning AV mechanisms oftentimes do not match the actual case counts in the respective training data. As such, probability estimates are poorly calibrated. We are expanding our framework from PAN 2020 to include Bayes factor scoring (BFS) and an uncertainty adaptation layer (UAL) to address both problems. Experiments with the 2020/21 PAN AV shared task data show that the proposed method significantly reduces sensitivities to topical variations and significantly improves the system's calibration.

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。

↑