Robustly Leveraging Prior Knowledge in Text Classification
This work addresses the robustness issue in leveraging prior knowledge for text classification, which is incremental as it builds on existing methods like generalized expectation criteria.
The paper tackles the problem of robustness when incorporating prior knowledge into text classification models, proposing three regularization terms that achieve remarkable improvements and greater robustness compared to baselines.
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.