An Attention Mechanism for Answer Selection Using a Combined Global and Local View
This work addresses answer selection for question answering systems, with incremental improvements in attention mechanisms.
The authors tackled answer selection in question answering by proposing a new attention mechanism that uses both global and local views of the answer, achieving state-of-the-art results on the InsuranceQA dataset.
We propose a new attention mechanism for neural based question answering, which depends on varying granularities of the input. Previous work focused on augmenting recurrent neural networks with simple attention mechanisms which are a function of the similarity between a question embedding and an answer embeddings across time. We extend this by making the attention mechanism dependent on a global embedding of the answer attained using a separate network. We evaluate our system on InsuranceQA, a large question answering dataset. Our model outperforms current state-of-the-art results on InsuranceQA. Further, we visualize which sections of text our attention mechanism focuses on, and explore its performance across different parameter settings.