Bidirectional Context-Aware Hierarchical Attention Network for Document Understanding
This work addresses a specific bottleneck in document understanding for NLP researchers, offering an incremental improvement over existing methods.
The paper tackled the limitation of Hierarchical Attention Networks (HAN) where sentences are encoded in isolation by proposing a bidirectional context-aware hierarchical attention network (CAHAN) that uses surrounding sentences as context, achieving improved performance on sentiment and topic classification datasets with a modest computational increase.
The Hierarchical Attention Network (HAN) has made great strides, but it suffers a major limitation: at level 1, each sentence is encoded in complete isolation. In this work, we propose and compare several modifications of HAN in which the sentence encoder is able to make context-aware attentional decisions (CAHAN). Furthermore, we propose a bidirectional document encoder that processes the document forwards and backwards, using the preceding and following sentences as context. Experiments on three large-scale sentiment and topic classification datasets show that the bidirectional version of CAHAN outperforms HAN everywhere, with only a modest increase in computation time. While results are promising, we expect the superiority of CAHAN to be even more evident on tasks requiring a deeper understanding of the input documents, such as abstractive summarization. Code is publicly available.