Less is More: Simplifying Feature Extractors Prevents Overfitting for Neural Discourse Parsing Models
share
Complex feature extractors are widely employed for text representation building. However, these complex feature extractors can lead to severe overfitting problems especially when the training datasets are small, which is especially the case for several discourse parsing tasks. Thus, we propose to remove additional feature extractors and only utilize self-attention mechanism to exploit pretrained neural language models in order to mitigate the overfitting problem. Experiments on three common discourse parsing tasks (News Discourse Profiling, Rhetorical Structure Theory based Discourse Parsing and Penn Discourse Treebank based Discourse Parsing) show that powered by recent pretrained language models, our simplied feature extractors obtain better generalizabilities and meanwhile achieve comparable or even better system performance. The simplified feature extractors have fewer learnable parameters and less processing time. Codes will be released and this simple yet effective model can serve as a better baseline for future research.
READ FULL TEXT