Explicit Utilization of General Knowledge in Machine Reading Comprehension
This addresses the gap between MRC models and humans in data efficiency and noise robustness, representing an incremental improvement by hybridizing neural networks with external knowledge.
The paper tackles the problem of Machine Reading Comprehension (MRC) models' data hunger and noise sensitivity by integrating general knowledge from WordNet into an end-to-end model called Knowledge Aided Reader (KAR), which achieves comparable performance to state-of-the-art models and significantly improves robustness to noise, especially outperforming them by a large margin when only 20%-80% of training data is available.
To bridge the gap between Machine Reading Comprehension (MRC) models and human beings, which is mainly reflected in the hunger for data and the robustness to noise, in this paper, we explore how to integrate the neural networks of MRC models with the general knowledge of human beings. On the one hand, we propose a data enrichment method, which uses WordNet to extract inter-word semantic connections as general knowledge from each given passage-question pair. On the other hand, we propose an end-to-end MRC model named as Knowledge Aided Reader (KAR), which explicitly uses the above extracted general knowledge to assist its attention mechanisms. Based on the data enrichment method, KAR is comparable in performance with the state-of-the-art MRC models, and significantly more robust to noise than them. When only a subset (20%-80%) of the training examples are available, KAR outperforms the state-of-the-art MRC models by a large margin, and is still reasonably robust to noise.