High-order Refining for End-to-end Chinese Semantic Role Labeling
This work addresses a specific bottleneck in Chinese semantic role labeling for NLP applications, representing an incremental improvement over existing graph-based models.
The paper tackles the problem of first-order limitations in end-to-end Chinese semantic role labeling by introducing a high-order refining mechanism that enables interaction between all predicate-argument pairs, achieving state-of-the-art results on Chinese SRL datasets like CoNLL09 and Universal Proposition Bank.
Current end-to-end semantic role labeling is mostly accomplished via graph-based neural models. However, these all are first-order models, where each decision for detecting any predicate-argument pair is made in isolation with local features. In this paper, we present a high-order refining mechanism to perform interaction between all predicate-argument pairs. Based on the baseline graph model, our high-order refining module learns higher-order features between all candidate pairs via attention calculation, which are later used to update the original token representations. After several iterations of refinement, the underlying token representations can be enriched with globally interacted features. Our high-order model achieves state-of-the-art results on Chinese SRL data, including CoNLL09 and Universal Proposition Bank, meanwhile relieving the long-range dependency issues.