Hierarchical Attention: What Really Counts in Various NLP Tasks
arXiv:1808.03728
Abstract
Attention mechanisms in sequence to sequence models have shown great ability and wonderful performance in various natural language processing (NLP) tasks, such as sentence embedding, text generation, machine translation, machine reading comprehension, etc. Unfortunately, existing attention mechanisms only learn either high-level or low-level features. In this paper, we think that the lack of hierarchical mechanisms is a bottleneck in improving the performance of the attention mechanisms, and propose a novel Hierarchical Attention Mechanism (Ham) based on the weighted sum of different layers of a multi-level attention. Ham achieves a state-of-the-art BLEU score of 0.26 on Chinese poem generation task and a nearly 6.5% averaged improvement compared with the existing machine reading comprehension models such as BIDAF and Match-LSTM. Furthermore, our experiments and theorems reveal that Ham has greater generalization and representation ability than existing attention mechanisms.
References in corpus (6)
- Efficient Estimation of Word Representations in Vector Space
- Bidirectional Attention Flow for Machine Comprehension
- A Structured Self-attentive Sentence Embedding
- Machine Comprehension Using Match-LSTM and Answer Pointer
- R: Reinforced Reader-Ranker for Open-Domain Question Answering
- DuReader: a Chinese Machine Reading Comprehension Dataset from Real-world Applications