Semantic Enhanced Text-to-SQL Parsing via Iteratively Learning Schema Linking Graph

Liu, Aiwei; Hu, Xuming; Li, Lin; Wen, Lijie

doi:10.1145/3534678.3539294

Cited by 14 publications

(7 citation statements)

References 34 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Schema-linking-based methods, such as RAT-SQL (Wang et al 2019), introduce a relation-aware transformer encoder to improve the joint encoding of a question and schema. Liu et al (2022) propose a similarity learning-based question-schema-alignment method to obtain a semantic schema-linking graph and observed how the pre-trained language model (PLM) embeddings for the schema items are affected. Zhao and Yang (2022) use the same words that appear in both the natural language statement and the table as weak supervised key points and design an interaction network to explore the correlation between the representations of the natural language statements and tables.…”

Section: Table-question Alignmentmentioning

confidence: 99%

A Table Question Alignment based Cell-Selection Method for Table-Text QA

Wu,

Xu,

F. Karlsson

et al. 2024

Journal of Natural Language Processing

View full text Add to dashboard Cite

Hybrid Question-Answering (HQA), which targets reasoning over tables and passages linked from table cells, has witnessed significant research in recent years. A common challenge in HQA and other passage-table QA datasets is that it is generally unrealistic to iterate over all table rows, columns, and linked passages to retrieve evidence. Such a challenge made it difficult for previous studies to show their reasoning ability in retrieving answers. To bridge this gap, we propose a novel Table-alignment-based Cell-selection and Reasoning model (TACR) for hybrid text and table QA, evaluated on the HybridQA and WikiTableQuestions datasets. In evidence retrieval, we design a table-question-alignment enhanced cell-selection method to retrieve finegrained evidence. In answer reasoning, we incorporate a QA module that treats the row containing selected cells as context. Experimental results over the HybridQA and WikiTableQuestions (WTQ) datasets show that TACR achieves state-of-the-art results on cell selection and outperforms fine-grained evidence retrieval baselines on HybridQA, while achieving competitive performance on WTQ. We also conducted a detailed analysis to demonstrate that being able to align questions to tables in the cell-selection stage can result in important gains from experiments of over 90% table row and column selection accuracy, meanwhile also improving output explainability.

show abstract

Section: Table-question Alignmentmentioning

confidence: 99%

A Table Question Alignment based Cell-Selection Method for Table-Text QA

Wu,

Xu,

F. Karlsson

et al. 2024

Journal of Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…Lastly, the performance gain of RAPL on cross-domain tasks is lower than that on in-domain tasks. An intriguing avenue for future research is to explore techniques for better performance on cross-domain tasks, e.g., data augmentation (Hu et al, 2023c) and structured knowledge guidance (Liu et al, 2022a;Ma et al, 2023a).…”

Section: Limitationsmentioning

confidence: 99%

RAPL: A Relation-Aware Prototype Learning Approach for Few-Shot Document-Level Relation Extraction

Meng,

Hu,

Liu

et al. 2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

How to identify semantic relations among entities in a document when only a few labeled documents are available? Few-shot documentlevel relation extraction (FSDLRE) is crucial for addressing the pervasive data scarcity problem in real-world scenarios. Metric-based meta-learning is an effective framework widely adopted for FSDLRE, which constructs class prototypes for classification. However, existing works often struggle to obtain class prototypes with accurate relational semantics: 1) To build prototype for a target relation type, they aggregate the representations of all entity pairs holding that relation, while these entity pairs may also hold other relations, thus disturbing the prototype. 2) They use a set of generic NOTA (none-of-the-above) prototypes across all tasks, neglecting that the NOTA semantics differs in tasks with different target relation types. In this paper, we propose a relation-aware prototype learning method for FSDLRE to strengthen the relational semantics of prototype representations. By judiciously leveraging the relation descriptions and realistic NOTA instances as guidance, our method effectively refines the relation prototypes and generates task-specific NOTA prototypes. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches by average 2.61% F 1 across various settings of two FSDLRE benchmarks. 1

show abstract

“…This formulation is related to incremental parsing, where a sentence is scanned from left-to-right and the structured is built incrementally by inserting a node or attaching an edge. Incremental parsers are widely used in semantic parsing (Zhou et al, 2016;Cheng et al, 2017;Guo and Lu, 2018;Naseem et al, 2019;Liu et al, 2022a) and syntactic parsing (Huang and Sagae, 2010;Dyer et al, 2015;Liu and Zhang, 2017), as they are computationally efficient, and can use machine learning to predict actions based on partially generated structures. Our feature fusion module can be viewed as the parser state as it carries the structural information and serves as a writable memory during the expansion step.…”

Section: Related Workmentioning

confidence: 99%

Scene Graph Modification as Incremental Structure Expanding

Hu¹,

Guo²,

Fu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

A scene graph is a semantic representation that expresses the objects, attributes, and relationships between objects in a scene. Scene graphs play an important role in many cross modality tasks, as they are able to capture the interactions between images and texts. In this paper, we focus on scene graph modification (SGM), where the system is required to learn how to update an existing scene graph based on a natural language query. Unlike previous approaches that rebuilt the entire scene graph, we frame SGM as a graph expansion task by introducing the incremental structure expanding (ISE). ISE constructs the target graph by incrementally expanding the source graph without changing the unmodified structure. Based on ISE, we further propose a model that iterates between nodes prediction and edges prediction, inferring more accurate and harmonious expansion decisions progressively. In addition, we construct a challenging dataset that contains more complicated queries and larger scene graphs than existing datasets. Experiments on four benchmarks demonstrate the effectiveness of our approach, which surpasses the previous state-of-the-art model by large margins. Source code and data are available 1 .1 https://github.com/THU-BPM/SGM * Equally Contributed.

show abstract

Semantic Enhanced Text-to-SQL Parsing via Iteratively Learning Schema Linking Graph

Cited by 14 publications

References 34 publications

A Table Question Alignment based Cell-Selection Method for Table-Text QA

A Table Question Alignment based Cell-Selection Method for Table-Text QA

RAPL: A Relation-Aware Prototype Learning Approach for Few-Shot Document-Level Relation Extraction

Scene Graph Modification as Incremental Structure Expanding

Contact Info

Product

Resources

About