ArcaneQA: Dynamic Program Induction and Contextualized Encoding for Knowledge Base Question Answering

04/17/2022
by   Yu Gu, et al.
0

Question answering on knowledge bases (KBQA) poses a unique challenge for semantic parsing research due to two intertwined factors: large search space and ambiguities in schema linking. The predominant ranking-based KBQA models, which rely on a candidate enumeration step to reduce the search space, struggle with flexibility and have impractical online running time. In this paper, we present ArcaneQA, a novel generation-based model that addresses both the large search space and schema linking in a unified framework with two mutually boosting ingredients: we use dynamic program induction to tackle the large search space and dynamic contextualized encoding to enhance schema linking. Experiment results on multiple popular KBQA datasets demonstrate the highly competitive performance of ArcaneQA in both effectiveness and efficiency.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/16/2021

Generative Relation Linking for Question Answering over Knowledge Bases

Relation linking is essential to enable question answering over knowledg...
research
04/27/2017

Question Answering on Knowledge Bases and Text using Universal Schema and Memory Networks

Existing question answering methods infer answers either from a knowledg...
research
07/06/2023

VisKoP: Visual Knowledge oriented Programming for Interactive Knowledge Base Question Answering

We present Visual Knowledge oriented Programming platform (VisKoP), a kn...
research
09/12/2022

Knowledge Base Question Answering: A Semantic Parsing Perspective

Recent advances in deep learning have greatly propelled the research on ...
research
10/12/2021

Program Transfer and Ontology Awareness for Semantic Parsing in KBQA

Semantic parsing in KBQA aims to parse natural language questions into l...
research
09/17/2021

RnG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering

Existing KBQA approaches, despite achieving strong performance on i.i.d....

Please sign up or login with your details

Forgot password? Click here to reset