arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2306.04544cs.CLcs.AI
收藏

用于标签精炼的对比自举方法

Contrastive Bootstrapping for Label Refinement

Peking University(北京大学);University of Southern California(南加州大学)

Shudi Hou, Yu Xia, Muhao Chen, Sujian Li

AI总结:

针对仅有粗粒度标签和粗到细映射的细粒度文本分类问题,提出一种轻量级对比聚类自举方法,从全局与局部视角迭代精炼段落标签,并在 NYT 和 20News 上显著超越现有方法。

摘要:

传统文本分类通常将文本归入预定义的粗粒度类别,由此产生的模型无法应对现实世界中为提供准确服务而周期性出现更细粒度类别的场景。在本工作中,我们研究了仅利用粗粒度类别标注和粗到细映射来完成细粒度分类的设定。我们提出一种轻量级的基于对比聚类的自举方法,以迭代方式精炼段落标签。在聚类过程中,该方法在映射指导下,从全局和局部两个视角拉开负向段落-原型对之间的距离。在 NYT 和 20News 上的实验表明,我们的方法大幅优于现有最先进方法。

英文摘要:

Traditional text classification typically categorizes texts into pre-defined coarse-grained classes, from which the produced models cannot handle the real-world scenario where finer categories emerge periodically for accurate services. In this work, we investigate the setting where fine-grained classification is done only using the annotation of coarse-grained categories and the coarse-to-fine mapping. We propose a lightweight contrastive clustering-based bootstrapping method to iteratively refine the labels of passages. During clustering, it pulls away negative passage-prototype pairs under the guidance of the mapping from both global and local perspectives. Experiments on NYT and 20News show that our method outperforms the state-of-the-art methods by a large margin.

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。

↑