Large Scale Question Paraphrase Retrieval with Smoothed Deep Metric Learning
Daniele Bonadiman, Anjishnu Kumar, Arpit Mittal · 2019
The goal of a Question Paraphrase Retrieval (QPR) system is to retrieve similar questions that result in the same answer as the original question.Such a system can be used to understand and answer rare and noisy reformulations of common questions by mapping them to a set of canonical forms.This task has large-scale applications for community Question Answering (cQA) and opendomain spoken language question-answering systems.In this paper, we describe a new QPR system implemented as a Neural Information Retrieval (NIR) system consisting of a neural network sentence encoder and an approximate k-Nearest Neighbour index for efficient vector retrieval.We also describe our mechanism to generate an annotated dataset for question paraphrase retrieval experiments automatically from question-answer logs via distant supervision.We show that the standard loss function in NIR, triplet loss, does not perform well with noisy labels.We propose the smoothed deep metric loss (SDML), and with our experiments on two QPR datasets we show that it significantly outperforms triplet loss in the noisy label setting.