Bridging Textual and Tabular Data for Cross-Domain Text-to-SQL Semantic Parsing (2020-11-01T00:00:00.000000Z)

TL;DR

This work presents BRIDGE, a powerful sequential architecture for modeling dependencies between natural language questions and relational databases in cross-DB semantic parsing that effectively captures the desired cross-modal dependencies and has the potential to generalize to more text-DB related tasks.

Abstract

We present BRIDGE, a powerful sequential architecture for modeling dependencies between natural language questions and relational databases in cross-DB semantic parsing. BRIDGE represents the question and DB schema in a tagged sequence where a subset of the fields are augmented with cell values mentioned in the question. The hybrid sequence is encoded by BERT with minimal subsequent layers and the text-DB contextualization is realized via the fine-tuned deep attention in BERT. Combined with a pointer-generator decoder with schema-consistency driven search space pruning, BRIDGE attained state-of-the-art performance on the well-studied Spider benchmark (65.5% dev, 59.2% test), despite being much simpler than most recently proposed models for this task. Our analysis shows that BRIDGE effectively captures the desired cross-modal dependencies and has the potential to generalize to more text-DB related tasks. Our model implementation is available at https://github.com/ salesforce/TabularSemanticParsing.

Authors

R. Socher

30 papers

Caiming Xiong

24 papers

Xi Victoria Lin

9 papers

TL;DR

Abstract

Authors

References66 items

Mention Extraction and Linking for SQL Query Generation

Compositional Generalization and Natural Language Variation: Can a Semantic Parsing Approach Handle Both?

SmBoP: Semi-autoregressive Bottom-up Semantic Parsing

Span-based Semantic Parsing for Compositional Generalization

Hybrid Ranking Network for Text-to-SQL

Compositional Generalization in Semantic Parsing: Pre-training vs. Specialized Architectures

An Empirical Study on Robustness to Spurious Correlations using Pre-trained Language Models

Photon: A Robust Cross-Domain Text-to-SQL System

Exploring Unexplored Generalization Challenges for Cross-Database Semantic Parsing

TaBERT: Pretraining for Joint Understanding of Textual and Tabular Data

An Imitation Game for Learning Semantic Parsers from User Interaction

Longformer: The Long-Document Transformer

RYANSQL: Recursively Applying Sketch-based Slot Fillings for Complex Text-to-SQL in Cross-Domain Databases

TaPas: Weakly Supervised Table Parsing via Pre-training

Bertrand-DR: Improving Text-to-SQL using a Discriminative Re-ranker

RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers

Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer

Content Enhanced BERT-based Text-to-SQL Generation

HuggingFace's Transformers: State-of-the-art Natural Language Processing

CoSQL: A Conversational Text-to-SQL Challenge Towards Cross-Domain Natural Language Interfaces to Databases

Editing-Based SQL Query Generation for Cross-Domain Context-Dependent Questions

Global Reasoning over Database Structures for Text-to-SQL Parsing

X-SQL: reinforce schema representation with context

RoBERTa: A Robustly Optimized BERT Pretraining Approach

SpanBERT: Improving Pre-training by Representing and Predicting Spans

Towards Comprehensive Description Generation from Factual Attribute-value Tables

A Multiscale Visualization of Attention in the Transformer Model

SParC: Cross-Domain Semantic Parsing in Context

Generating Logical Forms from Graph Representations of Text and Entities

Towards Complex Text-to-SQL in Cross-Domain Database with Intermediate Representation

Representing Schema Structure with Graph Neural Networks for Text-to-SQL Parsing

A Comprehensive Exploration on WikiSQL with Table-Aware Word Contextualization

Multi-Task Deep Neural Networks for Natural Language Understanding

SyntaxSQLNet: Syntax Tree Networks for Complex and Cross-Domain Text-to-SQL Task

Technologies

Execution-Guided Neural Program Decoding

Memory Augmented Policy Optimization for Program Synthesis and Semantic Parsing

Linguistically-Informed Self-Attention for Semantic Role Labeling

Self-Attention with Relative Position Representations

NL2Bash: A Corpus and Semantic Parser for Natural Language Interface to the Linux Operating System

Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning

Attention is All you Need

Get To The Point: Summarization with Pointer-Generator Networks

Incorporating Copying Mechanism in Sequence-to-Sequence Learning

Language to Logical Form with Neural Attention

Adam: A Method for Stochastic Optimization

Exploring the Limits

Learning to Map Sentences to Logical Form: Structured Classification with Probabilistic Categorial Grammars

Long Short-Term Memory

Learning to Parse Database Queries Using Inductive Logic Programming

Natural language interfaces to databases – an introduction

Expanding the Scope of the ATIS Task: The ATIS-3 Corpus

Natural language interfaces to databases

The ATIS Spoken Language Systems Pilot Corpus

Bridging textual and tabular data for cross-domain text-to-sql semantic parsing

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

What It Takes to Achieve 100 Percent Condition Accuracy on WikiSQL

A syntactic neural model for general-purpose code generation

Database Systems Design Implementation And Management

The original formulation

Database systems - design, implementation, and management (2. ed.)

boundaries can be detected within i − 2 to j + 2, otherwise the match is invalidated

People_ID (b) People.People_ID → Poker_Player.People_ID Figure A4: Visualization of attention in layer 5 between a pair of foreign keys

with the cell value “cat” ( s c ) and the matched sub-string is “cat” ( s m )

The 58th annual meeting of the Association for Computational Linguistics (ACL)

Specically

Field of Study

Venue Information

Name

Type

URL