Large Scale Question Paraphrase Retrieval with Smoothed Deep Metric Learning

05/29/2019
by   Daniele Bonadiman, et al.
0

The goal of a Question Paraphrase Retrieval (QPR) system is to retrieve equivalent 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 has large-scale applications for community Question Answering (cQA) and open-domain 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 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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/07/2018

Improving Retrieval-Based Question Answering with Deep Inference Models

Question answering is one of the most important and difficult applicatio...
research
08/20/2019

Noisy Corruption Detection

We answer a question of Alon, Mossel, and Pemantle about the corruption ...
research
05/22/2019

ANTIQUE: A Non-Factoid Question Answering Benchmark

Considering the widespread use of mobile and voice search, answer passag...
research
07/23/2014

Learning Rank Functionals: An Empirical Study

Ranking is a key aspect of many applications, such as information retrie...
research
10/05/2018

POIReviewQA: A Semantically Enriched POI Retrieval and Question Answering Dataset

Many services that perform information retrieval for Points of Interest ...
research
12/08/2019

Exploring the Ideal Depth of Neural Network when Predicting Question Deletion on Community Question Answering

In recent years, Community Question Answering (CQA) has emerged as a pop...
research
09/10/2023

Duplicate Question Retrieval and Confirmation Time Prediction in Software Communities

Community Question Answering (CQA) in different domains is growing at a ...

Please sign up or login with your details

Forgot password? Click here to reset