RAG-Instruct: Boosting LLMs with Diverse Retrieval-Augmented Instructions Article Swipe

PDF

Wanlong Liu , J Chen , Ke Ji , Li Zhou , Wenyu Chen , Benyou Wang ·

YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2501.00353

Retrieval-Augmented Generation (RAG) has emerged as a key paradigm for enhancing large language models (LLMs) by incorporating external knowledge. However, current RAG methods face two limitations: (1) they only cover limited RAG scenarios. (2) They suffer from limited task diversity due to the lack of a general RAG dataset. To address these limitations, we propose RAG-Instruct, a general method for synthesizing diverse and high-quality RAG instruction data based on any source corpus. Our approach leverages (1) five RAG paradigms, which encompass diverse query-document relationships, and (2) instruction simulation, which enhances instruction diversity and quality by utilizing the strengths of existing instruction datasets. Using this method, we construct a 40K instruction dataset from Wikipedia, comprehensively covering diverse RAG scenarios and tasks. Experiments demonstrate that RAG-Instruct effectively enhances LLMs' RAG capabilities, achieving strong zero-shot performance and significantly outperforming various RAG baselines across a diverse set of tasks. RAG-Instruct is publicly available at https://github.com/FreedomIntelligence/RAG-Instruct.

Related Topics

Boosting (Machine Learning)

Computer Science

Artificial Intelligence

Concepts

Boosting (machine learning) Computer science Artificial intelligence

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2501.00353
PDF: https://arxiv.org/pdf/2501.00353
OA Status: green
Related Works: 10
OpenAlex ID: https://openalex.org/W4406031136

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4406031136

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2501.00353

Digital Object Identifier
Title: RAG-Instruct: Boosting LLMs with Diverse Retrieval-Augmented Instructions

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2024

Year of publication
Publication date: 2024-12-31

Full publication date if available
Authors: Wanlong Liu, J Chen, Ke Ji, Li Zhou, Wenyu Chen, Benyou Wang

List of authors in order
Landing page: https://arxiv.org/abs/2501.00353

Publisher landing page
PDF URL: https://arxiv.org/pdf/2501.00353

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/2501.00353

Direct OA link when available
Concepts: Boosting (machine learning), Computer science, Artificial intelligence

Top concepts (fields/topics) attached by OpenAlex
Cited by: 0

Total citation count in OpenAlex
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4406031136
doi	https://doi.org/10.48550/arxiv.2501.00353
ids.doi	https://doi.org/10.48550/arxiv.2501.00353
ids.openalex	https://openalex.org/W4406031136
fwci
type	preprint
title	RAG-Instruct: Boosting LLMs with Diverse Retrieval-Augmented Instructions
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10181
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9947999715805054
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	Natural Language Processing Techniques
topics[1].id	https://openalex.org/T13523
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.9620000123977661
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1703
topics[1].subfield.display_name	Computational Theory and Mathematics
topics[1].display_name	Mathematics, Computing, and Information Processing
topics[2].id	https://openalex.org/T10601
topics[2].field.id	https://openalex.org/fields/17
topics[2].field.display_name	Computer Science
topics[2].score	0.9384999871253967
topics[2].domain.id	https://openalex.org/domains/3
topics[2].domain.display_name	Physical Sciences
topics[2].subfield.id	https://openalex.org/subfields/1707
topics[2].subfield.display_name	Computer Vision and Pattern Recognition
topics[2].display_name	Handwritten Text Recognition Techniques
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C46686674
concepts[0].level	2
concepts[0].score	0.9327405691146851
concepts[0].wikidata	https://www.wikidata.org/wiki/Q466303
concepts[0].display_name	Boosting (machine learning)
concepts[1].id	https://openalex.org/C41008148
concepts[1].level	0
concepts[1].score	0.4249733090400696
concepts[1].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[1].display_name	Computer science
concepts[2].id	https://openalex.org/C154945302
concepts[2].level	1
concepts[2].score	0.23839828372001648
concepts[2].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[2].display_name	Artificial intelligence
keywords[0].id	https://openalex.org/keywords/boosting
keywords[0].score	0.9327405691146851
keywords[0].display_name	Boosting (machine learning)
keywords[1].id	https://openalex.org/keywords/computer-science
keywords[1].score	0.4249733090400696
keywords[1].display_name	Computer science
keywords[2].id	https://openalex.org/keywords/artificial-intelligence
keywords[2].score	0.23839828372001648
keywords[2].display_name	Artificial intelligence
language	en
locations[0].id	pmh:oai:arXiv.org:2501.00353
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/2501.00353
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/2501.00353
locations[1].id	doi:10.48550/arxiv.2501.00353
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2501.00353
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5115756468
authorships[0].author.orcid
authorships[0].author.display_name	Wanlong Liu
authorships[0].author_position	first
authorships[0].raw_author_name	Liu, Wanlong
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5062345029
authorships[1].author.orcid
authorships[1].author.display_name	J Chen
authorships[1].author_position	middle
authorships[1].raw_author_name	Chen, Junying
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5102000879
authorships[2].author.orcid	https://orcid.org/0000-0003-0761-6484
authorships[2].author.display_name	Ke Ji
authorships[2].author_position	middle
authorships[2].raw_author_name	Ji, Ke
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5101515468
authorships[3].author.orcid	https://orcid.org/0000-0002-0300-0394
authorships[3].author.display_name	Li Zhou
authorships[3].author_position	middle
authorships[3].raw_author_name	Zhou, Li
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5100687323
authorships[4].author.orcid	https://orcid.org/0000-0002-9933-8014
authorships[4].author.display_name	Wenyu Chen
authorships[4].author_position	middle
authorships[4].raw_author_name	Chen, Wenyu
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5057282504
authorships[5].author.orcid	https://orcid.org/0000-0002-1501-9914
authorships[5].author.display_name	Benyou Wang
authorships[5].author_position	last
authorships[5].raw_author_name	Wang, Benyou
authorships[5].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/2501.00353
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-01-04T00:00:00
display_name	RAG-Instruct: Boosting LLMs with Diverse Retrieval-Augmented Instructions
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10181
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9947999715805054
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	Natural Language Processing Techniques
related_works	https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2125652721, https://openalex.org/W1540371141, https://openalex.org/W1549363203, https://openalex.org/W2147697413, https://openalex.org/W2154063878, https://openalex.org/W4231274751, https://openalex.org/W2556012038
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2501.00353
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/2501.00353
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/2501.00353
primary_location.id	pmh:oai:arXiv.org:2501.00353
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/2501.00353
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/2501.00353
publication_date	2024-12-31
publication_year	2024
referenced_works_count	0
abstract_inverted_index.a	6, 45, 56, 107, 140
abstract_inverted_index.To	49
abstract_inverted_index.as	5
abstract_inverted_index.at	149
abstract_inverted_index.by	15, 94
abstract_inverted_index.is	146
abstract_inverted_index.of	44, 98, 143
abstract_inverted_index.on	68
abstract_inverted_index.to	41
abstract_inverted_index.we	53, 105
abstract_inverted_index.(1)	26, 75
abstract_inverted_index.(2)	33, 85
abstract_inverted_index.40K	108
abstract_inverted_index.Our	72
abstract_inverted_index.RAG	21, 31, 47, 64, 77, 116, 127, 137
abstract_inverted_index.and	62, 84, 92, 118, 133
abstract_inverted_index.any	69
abstract_inverted_index.due	40
abstract_inverted_index.for	9, 59
abstract_inverted_index.has	3
abstract_inverted_index.key	7
abstract_inverted_index.set	142
abstract_inverted_index.the	42, 96
abstract_inverted_index.two	24
abstract_inverted_index.They	34
abstract_inverted_index.data	66
abstract_inverted_index.face	23
abstract_inverted_index.five	76
abstract_inverted_index.from	36, 111
abstract_inverted_index.lack	43
abstract_inverted_index.only	28
abstract_inverted_index.task	38
abstract_inverted_index.that	122
abstract_inverted_index.they	27
abstract_inverted_index.this	103
abstract_inverted_index.(RAG)	2
abstract_inverted_index.LLMs'	126
abstract_inverted_index.Using	102
abstract_inverted_index.based	67
abstract_inverted_index.cover	29
abstract_inverted_index.large	11
abstract_inverted_index.these	51
abstract_inverted_index.which	79, 88
abstract_inverted_index.(LLMs)	14
abstract_inverted_index.across	139
abstract_inverted_index.method	58
abstract_inverted_index.models	13
abstract_inverted_index.source	70
abstract_inverted_index.strong	130
abstract_inverted_index.suffer	35
abstract_inverted_index.tasks.	119, 144
abstract_inverted_index.address	50
abstract_inverted_index.corpus.	71
abstract_inverted_index.current	20
abstract_inverted_index.dataset	110
abstract_inverted_index.diverse	61, 81, 115, 141
abstract_inverted_index.emerged	4
abstract_inverted_index.general	46, 57
abstract_inverted_index.limited	30, 37
abstract_inverted_index.method,	104
abstract_inverted_index.methods	22
abstract_inverted_index.propose	54
abstract_inverted_index.quality	93
abstract_inverted_index.various	136
abstract_inverted_index.However,	19
abstract_inverted_index.approach	73
abstract_inverted_index.covering	114
abstract_inverted_index.dataset.	48
abstract_inverted_index.enhances	89, 125
abstract_inverted_index.existing	99
abstract_inverted_index.external	17
abstract_inverted_index.language	12
abstract_inverted_index.paradigm	8
abstract_inverted_index.publicly	147
abstract_inverted_index.achieving	129
abstract_inverted_index.available	148
abstract_inverted_index.baselines	138
abstract_inverted_index.construct	106
abstract_inverted_index.datasets.	101
abstract_inverted_index.diversity	39, 91
abstract_inverted_index.encompass	80
abstract_inverted_index.enhancing	10
abstract_inverted_index.leverages	74
abstract_inverted_index.scenarios	117
abstract_inverted_index.strengths	97
abstract_inverted_index.utilizing	95
abstract_inverted_index.zero-shot	131
abstract_inverted_index.Generation	1
abstract_inverted_index.Wikipedia,	112
abstract_inverted_index.knowledge.	18
abstract_inverted_index.paradigms,	78
abstract_inverted_index.scenarios.	32
abstract_inverted_index.Experiments	120
abstract_inverted_index.demonstrate	121
abstract_inverted_index.effectively	124
abstract_inverted_index.instruction	65, 86, 90, 100, 109
abstract_inverted_index.performance	132
abstract_inverted_index.simulation,	87
abstract_inverted_index.RAG-Instruct	123, 145
abstract_inverted_index.high-quality	63
abstract_inverted_index.limitations,	52
abstract_inverted_index.limitations:	25
abstract_inverted_index.synthesizing	60
abstract_inverted_index.RAG-Instruct,	55
abstract_inverted_index.capabilities,	128
abstract_inverted_index.incorporating	16
abstract_inverted_index.outperforming	135
abstract_inverted_index.significantly	134
abstract_inverted_index.query-document	82
abstract_inverted_index.relationships,	83
abstract_inverted_index.comprehensively	113
abstract_inverted_index.Retrieval-Augmented	0
abstract_inverted_index.https://github.com/FreedomIntelligence/RAG-Instruct.	150
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	6
citation_normalized_percentile