Unseen Entity Handling in Complex Question Answering over Knowledge Base via Language Generation

• DOI: 10.18653/v1/2021.findings-emnlp.50
• Publication Date: 2021-12-28T17:24:05Z
• Authors: Xin Huang; Jung-Jae Kim; Bowei Zou
• Keywords: Interpretability; Executable; Named graph; Knowledge graph; Base (topology)

Complex question answering over knowledge base remains as a challenging task because it involves reasoning multiple pieces of information, including intermediate entities/relations and other constraints. Previous methods simplify the SPARQL query into such forms list or graph, missing constraints "filter" "order_by", present models specialized for generating those simplified from given question. We instead introduce novel approach that directly generates an executable without simplification, addressing issue unseen entities. adapt large scale pre-trained encoder-decoder show our method significantly outperforms previous also has higher interpretability computational efficiency than methods.