XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQL Article Swipe

PDF

Yifu Liu , Yin Zhu , Yingqi Gao , Zhiling Luo , Xiaoxia Li , X. Shi , Yú Hónɡ , Jinyang Gao , Haijun Yu , Bolin Ding , Jingren Zhou ·

YOU? · · 2025 · Open Access ·

To leverage the advantages of LLM in addressing challenges in the Text-to-SQL task, we present XiYan-SQL, an innovative framework effectively generating and utilizing multiple SQL candidates. It consists of three components: 1) a Schema Filter module filtering and obtaining multiple relevant schemas; 2) a multi-generator ensemble approach generating multiple highquality and diverse SQL queries; 3) a selection model with a candidate reorganization strategy implemented to obtain the optimal SQL query. Specifically, for the multi-generator ensemble, we employ a multi-task fine-tuning strategy to enhance the capabilities of SQL generation models for the intrinsic alignment between SQL and text, and construct multiple generation models with distinct generation styles by fine-tuning across different SQL formats. The experimental results and comprehensive analysis demonstrate the effectiveness and robustness of our framework. Overall, XiYan-SQL achieves a new SOTA performance of 75.63% on the notable BIRD benchmark, surpassing all previous methods. It also attains SOTA performance on the Spider test set with an accuracy of 89.65%.

Related Topics

Pachinko (Novel)

Red Dragon (Novel)

Doctor Sleep (Novel)

Turtles All The Way Down (Novel)

Sophie's Choice (Novel)

The Lincoln Highway (Novel)

North And South (Gaskell Novel)

Thunderball (Novel)

Exodus (Uris Novel)

Possession (Byatt Novel)

The 100 (Novel Series)

Rivers Of London (Novel)

Concepts

No concepts available.

Metadata

Type: article
Language: en
Landing Page: http://arxiv.org/abs/2507.04701
PDF: https://arxiv.org/pdf/2507.04701
OA Status: green
OpenAlex ID: https://openalex.org/W4415348225

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4415348225

Canonical identifier for this work in OpenAlex
Title: XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQL

Work title
Type: article

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-07-07

Full publication date if available
Authors: Yifu Liu, Yin Zhu, Yingqi Gao, Zhiling Luo, Xiaoxia Li, X. Shi, Yú Hónɡ, Jinyang Gao, Haijun Yu, Bolin Ding, Jingren Zhou

List of authors in order
Landing page: https://arxiv.org/abs/2507.04701

Publisher landing page
PDF URL: https://arxiv.org/pdf/2507.04701

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2507.04701

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4415348225
doi
ids.openalex	https://openalex.org/W4415348225
fwci	0.0
type	article
title	XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQL
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T13734
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.5062000155448914
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1702
topics[0].subfield.display_name	Artificial Intelligence
topics[0].display_name	Advanced Computational Techniques and Applications
is_xpac	False
apc_list
apc_paid
language	en
locations[0].id	pmh:oai:arXiv.org:2507.04701
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2507.04701
locations[0].version	submittedVersion
locations[0].raw_type	text
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2507.04701
indexed_in	arxiv
authorships[0].author.id	https://openalex.org/A5075365578
authorships[0].author.orcid	https://orcid.org/0000-0002-2018-2572
authorships[0].author.display_name	Yifu Liu
authorships[0].author_position	first
authorships[0].raw_author_name	Liu, Yifu
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5100343741
authorships[1].author.orcid	https://orcid.org/0009-0002-1063-515X
authorships[1].author.display_name	Yin Zhu
authorships[1].author_position	middle
authorships[1].raw_author_name	Zhu, Yin
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5020118426
authorships[2].author.orcid	https://orcid.org/0000-0002-3247-0272
authorships[2].author.display_name	Yingqi Gao
authorships[2].author_position	middle
authorships[2].raw_author_name	Gao, Yingqi
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5020904548
authorships[3].author.orcid	https://orcid.org/0000-0002-0540-7307
authorships[3].author.display_name	Zhiling Luo
authorships[3].author_position	middle
authorships[3].raw_author_name	Luo, Zhiling
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5100348753
authorships[4].author.orcid	https://orcid.org/0000-0002-3510-2942
authorships[4].author.display_name	Xiaoxia Li
authorships[4].author_position	middle
authorships[4].raw_author_name	Li, Xiaoxia
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5091299418
authorships[5].author.orcid	https://orcid.org/0000-0001-9910-9345
authorships[5].author.display_name	X. Shi
authorships[5].author_position	middle
authorships[5].raw_author_name	Shi, Xiaorong
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5101834055
authorships[6].author.orcid	https://orcid.org/0000-0003-0738-5620
authorships[6].author.display_name	Yú Hónɡ
authorships[6].author_position	middle
authorships[6].raw_author_name	Hong, Yuntao
authorships[6].is_corresponding	False
authorships[7].author.id	https://openalex.org/A5101403129
authorships[7].author.orcid	https://orcid.org/0000-0002-2094-3554
authorships[7].author.display_name	Jinyang Gao
authorships[7].author_position	middle
authorships[7].raw_author_name	Gao, Jinyang
authorships[7].is_corresponding	False
authorships[8].author.id	https://openalex.org/A5022118078
authorships[8].author.orcid	https://orcid.org/0000-0002-3398-0880
authorships[8].author.display_name	Haijun Yu
authorships[8].author_position	middle
authorships[8].raw_author_name	Li, Yu
authorships[8].is_corresponding	False
authorships[9].author.id	https://openalex.org/A5040297543
authorships[9].author.orcid	https://orcid.org/0000-0003-1535-9692
authorships[9].author.display_name	Bolin Ding
authorships[9].author_position	middle
authorships[9].raw_author_name	Ding, Bolin
authorships[9].is_corresponding	False
authorships[10].author.id	https://openalex.org/A5057864403
authorships[10].author.orcid	https://orcid.org/0000-0002-4220-2634
authorships[10].author.display_name	Jingren Zhou
authorships[10].author_position	last
authorships[10].raw_author_name	Zhou, Jingren
authorships[10].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2507.04701
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-20T00:00:00
display_name	XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQL
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T04:12:42.849631
primary_topic.id	https://openalex.org/T13734
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.5062000155448914
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1702
primary_topic.subfield.display_name	Artificial Intelligence
primary_topic.display_name	Advanced Computational Techniques and Applications
cited_by_count	0
locations_count	1
best_oa_location.id	pmh:oai:arXiv.org:2507.04701
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2507.04701
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2507.04701
primary_location.id	pmh:oai:arXiv.org:2507.04701
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2507.04701
primary_location.version	submittedVersion
primary_location.raw_type	text
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2507.04701
publication_date	2025-07-07
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	32, 43, 55, 59, 77, 129
abstract_inverted_index.1)	31
abstract_inverted_index.2)	42
abstract_inverted_index.3)	54
abstract_inverted_index.It	26, 144
abstract_inverted_index.To	0
abstract_inverted_index.an	16, 155
abstract_inverted_index.by	106
abstract_inverted_index.in	6, 9
abstract_inverted_index.of	4, 28, 85, 123, 133, 157
abstract_inverted_index.on	135, 149
abstract_inverted_index.to	64, 81
abstract_inverted_index.we	13, 75
abstract_inverted_index.LLM	5
abstract_inverted_index.SQL	24, 52, 68, 86, 94, 110
abstract_inverted_index.The	112
abstract_inverted_index.all	141
abstract_inverted_index.and	21, 37, 50, 95, 97, 115, 121
abstract_inverted_index.for	71, 89
abstract_inverted_index.new	130
abstract_inverted_index.our	124
abstract_inverted_index.set	153
abstract_inverted_index.the	2, 10, 66, 72, 83, 90, 119, 136, 150
abstract_inverted_index.BIRD	138
abstract_inverted_index.SOTA	131, 147
abstract_inverted_index.also	145
abstract_inverted_index.test	152
abstract_inverted_index.with	58, 102, 154
abstract_inverted_index.model	57
abstract_inverted_index.task,	12
abstract_inverted_index.text,	96
abstract_inverted_index.three	29
abstract_inverted_index.75.63%	134
abstract_inverted_index.Filter	34
abstract_inverted_index.Schema	33
abstract_inverted_index.Spider	151
abstract_inverted_index.across	108
abstract_inverted_index.employ	76
abstract_inverted_index.models	88, 101
abstract_inverted_index.module	35
abstract_inverted_index.obtain	65
abstract_inverted_index.query.	69
abstract_inverted_index.styles	105
abstract_inverted_index.89.65%.	158
abstract_inverted_index.attains	146
abstract_inverted_index.between	93
abstract_inverted_index.diverse	51
abstract_inverted_index.enhance	82
abstract_inverted_index.notable	137
abstract_inverted_index.optimal	67
abstract_inverted_index.present	14
abstract_inverted_index.results	114
abstract_inverted_index.Overall,	126
abstract_inverted_index.accuracy	156
abstract_inverted_index.achieves	128
abstract_inverted_index.analysis	117
abstract_inverted_index.approach	46
abstract_inverted_index.consists	27
abstract_inverted_index.distinct	103
abstract_inverted_index.ensemble	45
abstract_inverted_index.formats.	111
abstract_inverted_index.leverage	1
abstract_inverted_index.methods.	143
abstract_inverted_index.multiple	23, 39, 48, 99
abstract_inverted_index.previous	142
abstract_inverted_index.queries;	53
abstract_inverted_index.relevant	40
abstract_inverted_index.schemas;	41
abstract_inverted_index.strategy	62, 80
abstract_inverted_index.XiYan-SQL	127
abstract_inverted_index.alignment	92
abstract_inverted_index.candidate	60
abstract_inverted_index.construct	98
abstract_inverted_index.different	109
abstract_inverted_index.ensemble,	74
abstract_inverted_index.filtering	36
abstract_inverted_index.framework	18
abstract_inverted_index.intrinsic	91
abstract_inverted_index.obtaining	38
abstract_inverted_index.selection	56
abstract_inverted_index.utilizing	22
abstract_inverted_index.XiYan-SQL,	15
abstract_inverted_index.addressing	7
abstract_inverted_index.advantages	3
abstract_inverted_index.benchmark,	139
abstract_inverted_index.challenges	8
abstract_inverted_index.framework.	125
abstract_inverted_index.generating	20, 47
abstract_inverted_index.generation	87, 100, 104
abstract_inverted_index.innovative	17
abstract_inverted_index.multi-task	78
abstract_inverted_index.robustness	122
abstract_inverted_index.surpassing	140
abstract_inverted_index.Text-to-SQL	11
abstract_inverted_index.candidates.	25
abstract_inverted_index.components:	30
abstract_inverted_index.demonstrate	118
abstract_inverted_index.effectively	19
abstract_inverted_index.fine-tuning	79, 107
abstract_inverted_index.highquality	49
abstract_inverted_index.implemented	63
abstract_inverted_index.performance	132, 148
abstract_inverted_index.capabilities	84
abstract_inverted_index.experimental	113
abstract_inverted_index.Specifically,	70
abstract_inverted_index.comprehensive	116
abstract_inverted_index.effectiveness	120
abstract_inverted_index.reorganization	61
abstract_inverted_index.multi-generator	44, 73
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	11
citation_normalized_percentile.value	0.22740161
citation_normalized_percentile.is_in_top_1_percent	False
citation_normalized_percentile.is_in_top_10_percent	True