Knowledge Informed Semantic Parsing for Conversational Question Answering

Abstract

Smart assistants are tasked to answer various questions regarding world knowledge. These questions range from retrieval of simple facts to retrieval of complex, multi-hops question followed by various operators (i.e., filter, argmax). Semantic parsing has emerged as the state-of-the-art for answering these kinds of questions by forming queries to extract information from knowledge bases (KBs). Specially, neural semantic parsers (NSPs) effectively translate natural questions to logical forms, which execute on KB and give desirable answers. Yet, NSPs suffer from nonexecutable logical forms for some instances in the generated logical forms might be missing due to the incompleteness of KBs. Intuitively, knowing the KB structure informs NSP with changes of the global logical forms structures with respect to changes in KB instances. In this work, we propose a novel knowledgeinformed decoder variant of NSP. We consider the conversational question answering settings, where a natural language query, its context and its final answers are available at training. Experimental results show that our method outperformed strong baselines by 1.8 F1 points overall across 10 types of questions of the CSQA dataset. Especially for the "Logical Reasoning" category, our model improves by 7 F1 points. Furthermore, our results are achieved with 90.3% fewer parameters, allowing faster training for large-scale datasets.

Structured Context and High-Coverage Grammar for Conversational Question Answering over Knowledge Graphs

“…This work was further extended in , which includes a graph attention network to exploit correlations between entities and predicates. Another related contemporaneous work is Thirukovalluru et al (2021), where the decoder is informed with entity embeddings coming from KG random walks.…”

Instilling Type Knowledge in Language Models via Multi-Task QA

“…Much like how knowledge acquisition in cognitive development progresses from recognizing concrete objects to gradually understanding their relations to one another (Lucariello et al, 1992), we aim to extend language models' existing rough understanding of entities (Heinzerling and Inui, 2021) to the types that govern how entities are related. Instilling type knowledge in multi-purpose models can improve performance in tasks like entity linking (Onoe and Durrett, 2020), question-answering (Févry et al, 2020a), and semantic parsing (Thirukovalluru et al, 2021).…”