Robustly Leveraging Prior Knowledge in Text Classification

TL;DR

This paper introduces three regularization techniques to enhance the robustness of text classification models by effectively leveraging prior knowledge, demonstrating significant improvements and stability across experiments.

Contribution

The paper proposes novel regularization methods on generalized expectation criteria to improve robustness in incorporating prior knowledge for text classification.

Findings

Proposed methods outperform baselines in accuracy.

Methods show increased robustness to knowledge variations.

Significant improvements across multiple datasets.

Abstract

Prior knowledge has been shown very useful to address many natural language processing tasks. Many approaches have been proposed to formalise a variety of knowledge, however, whether the proposed approach is robust or sensitive to the knowledge supplied to the model has rarely been discussed. In this paper, we propose three regularization terms on top of generalized expectation criteria, and conduct extensive experiments to justify the robustness of the proposed methods. Experimental results demonstrate that our proposed methods obtain remarkable improvements and are much more robust than baselines.