A Variational Approach for Mitigating Entity Bias in Relation Extraction
It addresses a critical generalization problem in relation extraction for NLP applications, though it appears incremental as it builds on an existing framework.
The paper tackled entity bias in relation extraction by adapting a Variational Information Bottleneck framework to compress entity-specific information, achieving state-of-the-art performance across general, financial, and biomedical domains in both in-domain and out-of-domain settings.
Mitigating entity bias is a critical challenge in Relation Extraction (RE), where models often rely excessively on entities, resulting in poor generalization. This paper presents a novel approach to address this issue by adapting a Variational Information Bottleneck (VIB) framework. Our method compresses entity-specific information while preserving task-relevant features. It achieves state-of-the-art performance on relation extraction datasets across general, financial, and biomedical domains, in both indomain (original test sets) and out-of-domain (modified test sets with type-constrained entity replacements) settings. Our approach offers a robust, interpretable, and theoretically grounded methodology.