Conversational Question Answering over Knowledge Graphs with Transformer and Graph Attention Networks

This paper addresses the task of (complex) conversational question answering over a knowledge graph. For this task, we propose LASAGNE (muLti-task semAntic parSing with trAnsformer and Graph atteNtion nEworks). It is the first approach, which employs a transformer architecture extended with Graph Attention Networks for multi-task neural semantic parsing. LASAGNE uses a transformer model for generating the base logical forms, while the Graph Attention model is used to exploit correlations between (entity) types and predicates to produce node representations. LASAGNE also includes a novel entity recognition module which detects, links, and ranks all relevant entities in the question context. We evaluate LASAGNE on a standard dataset for complex sequential question answering, on which it outperforms existing baselines averaged on all question types. Specifically, we show that LASAGNE improves the F1-score on eight out of ten question types; in some cases, the increase is more than 20% compared to state of the art (SotA).

[1]  Joelle Pineau,et al.  Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models , 2015, AAAI.

[2]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[3]  Roberto Cipolla,et al.  Multi-task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[4]  Chen Liang,et al.  Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision , 2016, ACL.

[5]  Ming-Wei Chang,et al.  The Value of Semantic Parse Labeling for Knowledge Base Question Answering , 2016, ACL.

[6]  Gerhard Weikum,et al.  Look before you Hop: Conversational Question Answering over Knowledge Graphs Using Judicious Context Expansion , 2019, CIKM.

[7]  Jens Lehmann,et al.  DBpedia - A large-scale, multilingual knowledge base extracted from Wikipedia , 2015, Semantic Web.

[8]  Pietro Liò,et al.  Graph Attention Networks , 2017, ICLR.

[9]  Mirella Lapata,et al.  Language to Logical Form with Neural Attention , 2016, ACL.

[10]  Johannes Hoffart,et al.  Evaluating the Impact of Knowledge Graph Context on Entity Disambiguation Models , 2020, CIKM.

[11]  Jens Lehmann,et al.  Why Reinvent the Wheel: Let's Build Question Answering Systems Together , 2018, WWW.

[12]  Akhilesh Vyas,et al.  Encoding Knowledge Graph Entity Aliases in Attentive Neural Network for Wikidata Entity Linking , 2019, WISE.

[13]  Jason Weston,et al.  Key-Value Memory Networks for Directly Reading Documents , 2016, EMNLP.

[14]  Nan Duan,et al.  Multi-Task Learning for Conversational Question Answering over a Large-Scale Knowledge Base , 2019, EMNLP.

[15]  Lihong Li,et al.  Neural Approaches to Conversational AI , 2019, Found. Trends Inf. Retr..

[16]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[17]  Ming Zhou,et al.  Dialog-to-Action: Conversational Question Answering Over a Large-Scale Knowledge Base , 2018, NeurIPS.

[18]  Hwee Tou Ng,et al.  A Generative Model for Parsing Natural Language to Meaning Representations , 2008, EMNLP.

[19]  Jeffrey Pennington,et al.  GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[20]  Mitesh M. Khapra,et al.  Complex Sequential Question Answering: Towards Learning to Converse Over Linked Question Answer Pairs with a Knowledge Graph , 2018, AAAI.

[21]  Le Song,et al.  Variational Reasoning for Question Answering with Knowledge Graph , 2017, AAAI.

[22]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[23]  Ralph Ewerth,et al.  MLM: A Benchmark Dataset for Multitask Learning with Multiple Languages and Modalities , 2020, CIKM.

[24]  Gerhard Weikum,et al.  WWW 2007 / Track: Semantic Web Session: Ontologies ABSTRACT YAGO: A Core of Semantic Knowledge , 2022 .

[25]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..