HIE-SQL: History Information Enhanced Network for Context-Dependent Text-to-SQL Semantic Parsing

Zheng, Yanzhao; Wang, Haibin; Dong, Biao; Wang, Xingjun; Li, Changshan

doi:10.18653/v1/2022.findings-acl.236

Cited by 14 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…EditSQL (Zhang et al, 2019) 39.9 12.3 GAZP (Zhong et al, 2020) 42.0 12.3 IGSQL (Cai and Wan, 2020) 44.1 15.8 RichContext 41.0 14.0 IST-SQL 44.4 14.7 R 2 SQL (Hui et al, 2021) 45.7 19.5 DELTA ♥ 51.7 21.5 SCORE ♦ (Yu et al, 2021b) 52.1 22.0 PICARD † (Scholak et al, 2021) 56.9 24.2 HIE-SQL ♦ (Zheng et al, 2022) 56.4 28.7…”

Section: Resultsmentioning

confidence: 99%

“…CQR-SQL also surpasses PICARD (Scholak et al, 2021) which is based on the super large pre-trained models T5-3B by +1.5% QM scores and +5.2% IM scores on COSQL. Compared with recent HIE-SQL (Zheng et al, 2022) who leverages rich representations of previous SQL queries from a pre-trained SQL language model, our CQR-SQL can also achieve significant improvements.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

CQR-SQL: Conversational Question Reformulation Enhanced Context-Dependent Text-to-SQL Parsers

Xiao¹,

Chai²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Context-dependent text-to-SQL is the task of translating multi-turn questions into databaserelated SQL queries. Existing methods typically focus on making full use of history context or previously predicted SQL for currently SQL parsing, while neglecting to explicitly comprehend the schema and conversational dependency, such as co-reference, ellipsis and user focus change. In this paper, we propose CQR-SQL, which uses auxiliary Conversational Question Reformulation (CQR) learning to explicitly exploit schema and decouple contextual dependency for SQL parsing. Specifically, we first present a schema enhanced recursive CQR method to produce domain-relevant selfcontained questions. Secondly, we train CQR-SQL models to map the semantics of multi-turn questions and auxiliary self-contained questions into the same latent space through schema grounding consistency task and tree-structured SQL parsing consistency task, which enhances the abilities of SQL parsing by adequately contextual understanding. At the time of writing, our CQR-SQL achieves new state-of-the-art results on two context-dependent text-to-SQL benchmarks SPARC and COSQL.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

CQR-SQL: Conversational Question Reformulation Enhanced Context-Dependent Text-to-SQL Parsers

Xiao¹,

Chai²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In this scenario, most SOTA parsers (Guo et al 2019;Wang et al 2020a;Cao et al 2021) apply top-down grammar-based decoding, which is consistent with our work. The last one is a conversational task, such as SPARC (Yu et al 2019b) and COSQL (Yu et al 2019a), forcing the parser to learn to consider the context information when generating SQL in a multi-turn dialogue (Zhang et al 2019;Zheng et al 2022). SSL in Semantic Parsing Multiple classical methods have been applied to address the challenge of lack of annotation for semantic parsing, such as SVM (Kate and Mooney 2007), self-training (Goldwasser et al 2011), dual learning (Chen et al 2021b), auto-encoder (Yin et al 2018), and mean-teacher (Wang et al 2020b).…”

Section: Related Workmentioning

confidence: 99%

Learn from Yesterday: A Semi-supervised Continual Learning Method for Supervision-Limited Text-to-SQL Task Streams

Chen¹,

Guo²,

et al. 2023

AAAI

View full text Add to dashboard Cite

Conventional text-to-SQL studies are limited to a single task with a fixed-size training and test set. When confronted with a stream of tasks common in real-world applications, existing methods struggle with the problems of insufficient supervised data and high retraining costs. The former tends to cause overfitting on unseen databases for the new task, while the latter makes a full review of instances from past tasks impractical for the model, resulting in forgetting of learned SQL structures and database schemas. To address the problems, this paper proposes integrating semi-supervised learning (SSL) and continual learning (CL) in a stream of text-to-SQL tasks and offers two promising solutions in turn. The first solution Vanilla is to perform self-training, augmenting the supervised training data with predicted pseudo-labeled instances of the current task, while replacing the full volume retraining with episodic memory replay to balance the training efficiency with the performance of previous tasks. The improved solution SFNet takes advantage of the intrinsic connection between CL and SSL. It uses in-memory past information to help current SSL, while adding high-quality pseudo instances in memory to improve future replay. The experiments on two datasets shows that SFNet outperforms the widely-used SSL-only and CL-only baselines on multiple metrics.

show abstract

“…In prior research in the field of multi-turn text-to-SQL, the primary emphasis has been on harnessing contextual information [8]- [10]. In a multi-turn text-to-SQL task, models are confronted with the challenge of concurrently handling both relational modeling and contextual modeling.…”

Section: Introductionmentioning

confidence: 99%

“…Several prior studies [9], [11]- [13] have employed neural network encoders that con- catenate the current question, question context, and schema. Concurrently, a number of approaches have directly incorporated historically generated SQL queries [8], [14]- [16] to aid the model in SQL parsing for the present question. However, these methods have tended to overlook the exploration of the wealth of structural information embedded within the dataset.…”

Section: Introductionmentioning

confidence: 99%

SE-HCL: Schema Enhanced Hybrid Curriculum Learning for Multi-Turn Text-to-SQL

Zhang,

Zhou,

Huang

2024

IEEE Access

View full text Add to dashboard Cite

Existing multi-turn Text-to-SQL approaches, mainly use data in a randomized order when training the model, ignoring the rich structural information contained in the dialog and schema. In this paper, we propose to use curriculum learning (CL) to better leverage the curriculum structure of schema, query, and dialog for multi-turn question-query pairs. We design a model-agnostic framework named Schema Enhanced Hybrid Curriculum Learning (SE-HCL) for multi-turn Text-to-SQL to help the models gain a full contextual semantic understanding. Concretely, We measure the difficulty of the data from both a structural and model perspective. In terms of data structure, we mainly consider the turns of the question and the complexity of the schema and SQL query. Accordingly, we designed a data course module to dynamically adjust the difficulty of the data based on the convergence of the model and the schema enhancement method we designed. In terms of the model, we propose a scoring module that will judge the difficulty of a problem based on whether the model could solve the question effectively. Finally, we will consider both aspects and design a hybrid curriculum to determine the flow of model training. Our experiments show that our proposed method improves SQL-generated performance over previous state-of-the-art models on SparC and CoSQL, especially for hard and long-turn questions.

show abstract

HIE-SQL: History Information Enhanced Network for Context-Dependent Text-to-SQL Semantic Parsing

Cited by 14 publications

References 0 publications

CQR-SQL: Conversational Question Reformulation Enhanced Context-Dependent Text-to-SQL Parsers

CQR-SQL: Conversational Question Reformulation Enhanced Context-Dependent Text-to-SQL Parsers

Learn from Yesterday: A Semi-supervised Continual Learning Method for Supervision-Limited Text-to-SQL Task Streams

SE-HCL: Schema Enhanced Hybrid Curriculum Learning for Multi-Turn Text-to-SQL

Contact Info

Product

Resources

About