Multi-task Sentence Encoding Model for Semantic Retrieval in Question Answering Systems

TL;DR

This paper introduces a multi-task sentence encoding model for semantic retrieval in question answering systems, combining graph-based sentence relations and multi-task learning to improve matching accuracy and retrieval speed.

Contribution

The paper presents a novel multi-task sentence encoding model that integrates graph relations and applies to sentence matching and intent classification in QA systems.

Findings

Outperforms existing sentence matching models in accuracy

Enables fast retrieval using Approximate Nearest Neighbor technology

Effective in real-time question answering scenarios

Abstract

Question Answering (QA) systems are used to provide proper responses to users' questions automatically. Sentence matching is an essential task in the QA systems and is usually reformulated as a Paraphrase Identification (PI) problem. Given a question, the aim of the task is to find the most similar question from a QA knowledge base. In this paper, we propose a Multi-task Sentence Encoding Model (MSEM) for the PI problem, wherein a connected graph is employed to depict the relation between sentences, and a multi-task learning model is applied to address both the sentence matching and sentence intent classification problem. In addition, we implement a general semantic retrieval framework that combines our proposed model and the Approximate Nearest Neighbor (ANN) technology, which enables us to find the most similar question from all available candidates very quickly during online serving. The experiments show the superiority of our proposed method as compared with the existing sentence matching models.