A Generic Method for Fine-grained Category Discovery in Natural Language Texts

TL;DR

This paper presents a novel contrastive learning method that improves fine-grained category discovery in texts by leveraging semantic similarities in a logarithmic space, with a centroid inference mechanism for real-time detection, outperforming existing methods.

Contribution

Introduces a new objective function guiding sample clustering based on semantic similarities in a logarithmic space and a centroid inference mechanism for real-time applications.

Findings

Outperforms state-of-the-art in accuracy, ARI, NMI

Validated on three benchmark tasks

Theoretically justified and empirically confirmed

Abstract

Fine-grained category discovery using only coarse-grained supervision is a cost-effective yet challenging task. Previous training methods focus on aligning query samples with positive samples and distancing them from negatives. They often neglect intra-category and inter-category semantic similarities of fine-grained categories when navigating sample distributions in the embedding space. Furthermore, some evaluation techniques that rely on pre-collected test samples are inadequate for real-time applications. To address these shortcomings, we introduce a method that successfully detects fine-grained clusters of semantically similar texts guided by a novel objective function. The method uses semantic similarities in a logarithmic space to guide sample distributions in the Euclidean space and to form distinct clusters that represent fine-grained categories. We also propose a centroid inference mechanism to support real-time applications. The efficacy of the method is both theoretically justified and empirically confirmed on three benchmark tasks. The proposed objective function is integrated in multiple contrastive learning based neural models. Its results surpass existing state-of-the-art approaches in terms of Accuracy, Adjusted Rand Index and Normalized Mutual Information of the detected fine-grained categories. Code and data will be available at Code and data are publicly available at https://github.com/changtianluckyforever/F-grained-STAR.