Compositional Semantic Parsing on Semi-Structured Tables

Abstract

Two important aspects of semantic parsing for question answering are the breadth of the knowledge source and the depth of logical compositionality. While existing work trades off one aspect for another, this paper simultaneously makes progress on both fronts through a new task: answering complex questions on semi-structured tables using question-answer pairs as supervision. The central challenge arises from two compounding factors: the broader domain results in an open-ended set of relations, and the deeper compositionality results in a combinatorial explosion in the space of logical forms. We propose a logical-form driven parsing algorithm guided by strong typing constraints and show that it obtains significant improvements over natural baselines. For evaluation, we created a new dataset of 22,033 complex questions on Wikipedia tables, which is made publicly available.

Cited by in corpus (49)

• Axiomatic Attribution for Deep Networks
• Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence
• TabFact: A Large-scale Dataset for Table-based Fact Verification
• A Syntactic Neural Model for General-Purpose Code Generation
• Neural Programmer: Inducing Latent Programs with Gradient Descent
• GraPPa: Grammar-Augmented Pre-Training for Table Semantic Parsing
• Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision
• From Language to Programs: Bridging Reinforcement Learning and Maximum Marginal Likelihood
• Neural Enquirer: Learning to Query Tables with Natural Language
• SyntaxSQLNet: Syntax Tree Networks for Complex and Cross-DomainText-to-SQL Task
• Simpler Context-Dependent Logical Forms via Model Projections
• A Transfer-Learnable Natural Language Interface for Databases
• Estimate and Replace: A Novel Approach to Integrating Deep Neural Networks with Existing Applications
• Learning Structured Natural Language Representations for Semantic Parsing
• Adversarial TableQA: Attention Supervision for Question Answering on Tables
• Answering Complex Questions Using Open Information Extraction
• Alignment-based compositional semantics for instruction following
• Semantic Parsing with Syntax- and Table-Aware SQL Generation
• TabMCQ: A Dataset of General Knowledge Tables and Multiple-choice Questions
• Neural network gradient-based learning of black-box function interfaces
• Neural Semantic Parsing in Low-Resource Settings with Back-Translation and Meta-Learning
• Exploring Neural Models for Parsing Natural Language into First-Order Logic
• Interactive Task and Concept Learning from Natural Language Instructions and GUI Demonstrations
• Representing Meaning with a Combination of Logical and Distributional Models
• Macro Grammars and Holistic Triggering for Efficient Semantic Parsing
• Coupling Distributed and Symbolic Execution for Natural Language Queries
• Neural Multi-Step Reasoning for Question Answering on Semi-Structured Tables
• Question Generation from SQL Queries Improves Neural Semantic Parsing
• ManyModalQA: Modality Disambiguation and QA over Diverse Inputs
• Generating Semantically Valid Adversarial Questions for TableQA
• Translating Natural Language to SQL using Pointer-Generator Networks and How Decoding Order Matters
• Transferable Natural Language Interface to Structured Queries aided by Adversarial Generation
• Effective Search of Logical Forms for Weakly Supervised Knowledge-Based Question Answering
• Neural Compositional Denotational Semantics for Question Answering
• CLTR: An End-to-End, Transformer-Based System for Cell Level Table Retrieval and Table Question Answering
• Web Question Answering with Neurosymbolic Program Synthesis
• Learning Executable Semantic Parsers for Natural Language Understanding
• Question Answering via Web Extracted Tables and Pipelined Models
• Learning Algebraic Recombination for Compositional Generalization
• Learning to Generate Structured Queries from Natural Language with Indirect Supervision
• FANDA: A Novel Approach to Perform Follow-up Query Analysis
• Learning to Learn Semantic Parsers from Natural Language Supervision
• Symbolic Priors for RNN-based Semantic Parsing
• Mapping Natural Language Commands to Web Elements
• Logical Parsing from Natural Language Based on a Neural Translation Model
• Bew: Towards Answering Business-Entity-Related Web Questions
• Content-Based Table Retrieval for Web Queries
• Context Dependent Semantic Parsing: A Survey