arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.03202cs.CL
收藏

一种用于多标签文本分类的去偏最近邻框架

A Debiased Nearest Neighbors Framework for Multi-Label Text Classification

State Key Laboratory for Novel Software Technology, Nanjing University(南京大学计算机软件新技术国家重点实验室)

Zifeng Cheng, Zhiwei Jiang, Yafeng Yin, Zhaoling Chen, Cong Wang, Shiping Ge, Qiguo Huang, Qing Gu

AI总结:

本文针对多标签文本分类中嵌入对齐和置信度估计两类偏差,提出去偏最近邻框架DENN,通过去偏对比学习和去偏置信度估计提升预测性能,在四个基准数据集上验证有效且不增加额外参数。

摘要:

多标签文本分类(MLTC)是一项实用但具有挑战性的任务,涉及为每个文档分配多个非互斥标签。以往研究主要关注通过引入特殊标注方案、设计特定模型结构或添加辅助任务来捕获标签相关性以辅助标签预测。近年来,$k$ 最近邻($k$NN)框架展现出潜力,它通过检索已标注样本作为参考,在嵌入空间中挖掘标签共现信息。然而,两个关键偏差,即嵌入对齐偏差和置信度估计偏差,常常被忽视,从而对预测性能产生不利影响。在本文中,我们提出了一种用于 MLTC 的去偏最近邻(DENN)框架,专门设计用于缓解这些偏差。针对嵌入对齐偏差,我们提出了一种去偏对比学习策略,增强邻居在标签共现上的一致性。针对置信度估计偏差,我们提出了一种去偏置信度估计策略,改进来自 $k$NN 和归纳式二分类的预测的自适应组合。在四个公开基准数据集(即 AAPD、RCV1-V2、Amazon-531 和 EUR-LEX57K)上进行的大量实验证明了我们所提方法的有效性。此外,我们的方法没有引入任何额外参数。

英文摘要:

Multi-Label Text Classification (MLTC) is a practical yet challenging task that involves assigning multiple non-exclusive labels to each document. Previous studies primarily focus on capturing label correlations to assist label prediction by introducing special labeling schemes, designing specific model structures, or adding auxiliary tasks. Recently, the $k$ Nearest Neighbor ($k$NN) framework has shown promise by retrieving labeled samples as references to mine label co-occurrence information in the embedding space. However, two critical biases, namely embedding alignment bias and confidence estimation bias, are often overlooked, adversely affecting prediction performance. In this paper, we introduce a DEbiased Nearest Neighbors (DENN) framework for MLTC, specifically designed to mitigate these biases. To address embedding alignment bias, we propose a debiased contrastive learning strategy, enhancing neighbor consistency on label co-occurrence. For confidence estimation bias, we present a debiased confidence estimation strategy, improving the adaptive combination of predictions from $k$NN and inductive binary classifications. Extensive experiments conducted on four public benchmark datasets (i.e., AAPD, RCV1-V2, Amazon-531, and EUR-LEX57K) showcase the effectiveness of our proposed method. Besides, our method does not introduce any extra parameters.

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。

↑