Semi-supervised Question Retrieval with Gated Convolutions

TL;DR

This paper introduces a semi-supervised gated convolutional model for question retrieval that effectively captures semantic similarity despite limited annotations, outperforming traditional IR methods and other neural networks.

Contribution

It proposes a novel gated convolutional approach with pre-training and fine-tuning for improved semantic question matching.

Findings

Significant performance improvements over IR baseline.

Outperforms CNNs, LSTMs, and GRUs in question retrieval.

Effective use of limited annotations through pre-training and fine-tuning.

Abstract

Question answering forums are rapidly growing in size with no effective automated ability to refer to and reuse answers already available for previous posted questions. In this paper, we develop a methodology for finding semantically related questions. The task is difficult since 1) key pieces of information are often buried in extraneous details in the question body and 2) available annotations on similar questions are scarce and fragmented. We design a recurrent and convolutional model (gated convolution) to effectively map questions to their semantic representations. The models are pre-trained within an encoder-decoder framework (from body to title) on the basis of the entire raw corpus, and fine-tuned discriminatively from limited annotations. Our evaluation demonstrates that our model yields substantial gains over a standard IR baseline and various neural network architectures (including CNNs, LSTMs and GRUs).