Improving Continual Relation Extraction by Distinguishing Analogous Semantics
This work addresses a specific bottleneck in continual learning for relation extraction, making it incremental for researchers in that domain.
The paper tackles the problem of overfitting in continual relation extraction caused by replaying stored samples, particularly for analogous relations, and proposes a model with memory-insensitive prototypes and augmentation that shows effectiveness in distinguishing analogous relations and overcoming overfitting.
Continual relation extraction (RE) aims to learn constantly emerging relations while avoiding forgetting the learned relations. Existing works store a small number of typical samples to re-train the model for alleviating forgetting. However, repeatedly replaying these samples may cause the overfitting problem. We conduct an empirical study on existing works and observe that their performance is severely affected by analogous relations. To address this issue, we propose a novel continual extraction model for analogous relations. Specifically, we design memory-insensitive relation prototypes and memory augmentation to overcome the overfitting problem. We also introduce integrated training and focal knowledge distillation to enhance the performance on analogous relations. Experimental results show the superiority of our model and demonstrate its effectiveness in distinguishing analogous relations and overcoming overfitting.