Zero-shot Text-to-SQL Learning with Auxiliary Task

@inproceedings{Chang2020ZeroshotTL,
  title={Zero-shot Text-to-SQL Learning with Auxiliary Task},
  author={Shuaichen Chang and Pengfei Liu and Yun Tang and Jing Huang and Xiaodong He and Bowen Zhou},
  booktitle={AAAI},
  year={2020}
}
Recent years have seen great success in the use of neural seq2seq models on the text-to-SQL task. [...] Key Result Compared to a strong baseline coarse-to-fine model, our models improve over the baseline by more than 3% absolute in accuracy on the whole dataset. More interestingly, on a zero-shot subset test of WikiSQL, our models achieve 5% absolute accuracy gain over the baseline, clearly demonstrating its superior generalizability.Expand
Leveraging Table Content for Zero-shot Text-to-SQL with Meta-Learning
TLDR
A new approach for the zero-shot text-to-SQL task which does not rely on any additional manual annotations is proposed and achieves significant improvements on both datasets. Expand
Translating Natural Language Questions to SQL Queries
Sequence-to-sequence models have performed well at the Text-to-SQL task on datasets such as WikiSQL. However, most prior work does not examine generalizability of the models to unfamiliar tableExpand
Hybrid Ranking Network for Text-to-SQL
TLDR
A neat approach to leverage pre-trained language models in Text-to-SQL called Hybrid Ranking Network (HydraNet) which breaks down the problem into column-wise ranking and decoding and finally assembles the column- wise outputs into a SQL query by straightforward rules. Expand
Neural Approaches for Natural Language Interfaces to Databases: A Survey
TLDR
This survey focuses on the key design decisions behind current state of the art neural approaches, which are group into encoder and decoder improvements, and highlights the three most important directions, namely linking question tokens to database schema elements, better architectures for encoding the textual query taking into account the schema, and improved generation of structured queries using autoregressive neural models. Expand
Multi-Task Learning in Natural Language Processing: An Overview
TLDR
An overview of the use of MTL in NLP tasks is given and optimization techniques on loss construction, data sampling, and task scheduling to properly train a multi-task model are presented. Expand
Conversational semantic parsing over tables by decoupling and grouping actions
TLDR
A new approach for answering a series of simple but inter-related questions over a table by mapping them to the corresponding SQLs by following the sequence-to-action paradigm and contributing an improved action space and a simplified scoring model. Expand
AutoQA: From Databases to Q&A Semantic Parsers with Only Synthetic Training Data
TLDR
The proposed AutoQA, a methodology and toolkit to generate semantic parsers that answer questions on databases, with no manual effort, is applied to the Schema2QA dataset and achieves an average logical form accuracy of 62.9% when tested on natural questions. Expand
Few-Shot Semantic Parsing for New Predicates
TLDR
This work proposed to apply a designated meta-learning method to train the model, regularize attention scores with alignment statistics, and apply a smoothing technique in pretraining, which consistently outperforms all the baselines in both one and two-shot settings. Expand
SQL Generation from Natural Language: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules
TLDR
This study presents a Sequence-to-Sequence (Seq2Seq) parsing model for the NL to SQL task, powered by the Transformers Architecture exploring the two Language Models (LM): Text-To-Text Transfer Transformer (T5) and the Multilingual pre-trained Text- to-Text Trans transformer (mT5). Expand
Semi-Automatic Construction of Text-to-SQL Data for Domain Transfer
Strong and affordable in-domain data is a desirable asset when transferring trained semantic parsers to novel domains. As previous methods for semi-automatically constructing such data cannot handleExpand
...
1
2
...

References

SHOWING 1-10 OF 28 REFERENCES
A Comprehensive Exploration on WikiSQL with Table-Aware Word Contextualization
TLDR
SQLova is presented, the first Natural-language-to-SQL (NL2SQL) model to achieve human performance in WikiSQL dataset, and it is shown that the model's performance is near the upper bound in Wiki SQL, where it is observed that a large portion of the evaluation errors are due to wrong annotations. Expand
Natural Language to Structured Query Generation via Meta-Learning
TLDR
This work explores a different learning protocol that treats each example as a unique pseudo-task, by reducing the original learning problem to a few-shot meta-learning scenario with the help of a domain-dependent relevance function. Expand
IncSQL: Training Incremental Text-to-SQL Parsers with Non-Deterministic Oracles
TLDR
A sequence-to-action parsing approach for the natural language to SQL task that incrementally fills the slots of a SQL query with feasible actions from a pre-defined inventory and sets a new state-of-the-art performance at an execution accuracy of 87.1%. Expand
Improving Text-to-SQL Evaluation Methodology
TLDR
It is shown that the current division of data into training and test sets measures robustness to variations in the way questions are asked, but only partially tests how well systems generalize to new queries, and proposes a complementary dataset split for evaluation of future work. Expand
Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
TLDR
This work proposes Seq2 SQL, a deep neural network for translating natural language questions to corresponding SQL queries, and releases WikiSQL, a dataset of 80654 hand-annotated examples of questions and SQL queries distributed across 24241 tables fromWikipedia that is an order of magnitude larger than comparable datasets. Expand
SQLNet: Generating Structured Queries From Natural Language Without Reinforcement Learning
TLDR
A sketch-based approach where the sketch contains a dependency graph, so that one prediction can be done by taking into consideration only the previous predictions that it depends on, and it is shown that SQLNet can outperform the prior art by 9% to 13% on the WikiSQL task. Expand
Robust Text-to-SQL Generation with Execution-Guided Decoding
TLDR
A new mechanism, execution guidance, is introduced to leverage the semantics of SQL, which detects and excludes faulty programs during the decoding procedure by conditioning on the execution of partially generated program. Expand
Learning a Natural Language Interface with Neural Programmer
TLDR
This paper presents the first weakly supervised, end-to-end neural network model to induce such programs on a real-world dataset, and enhances the objective function of Neural Programmer, a neural network with built-in discrete operations, and applies it on WikiTableQuestions, a natural language question-answering dataset. Expand
TypeSQL: Knowledge-Based Type-Aware Neural Text-to-SQL Generation
TLDR
This paper presents a novel approach TypeSQL which formats the problem as a slot filling task in a more reasonable way and utilizes type information to better understand rare entities and numbers in the questions. Expand
Bidirectional Attention Flow for Machine Comprehension
TLDR
The BIDAF network is introduced, a multi-stage hierarchical process that represents the context at different levels of granularity and uses bi-directional attention flow mechanism to obtain a query-aware context representation without early summarization. Expand
...
1
2
3
...