Large Language Model Can Be a Foundation for Hidden Rationale-Based Retrieval Article Swipe

PDF

Luo Ji , Fulai Guo , Teng Chen , Qing Gu , Xiaoyu Wang , Ningyuan Xi , Yihong Wang , Peng Yu , Yue Zhao , Hongyang Lei , Zhonglin Jiang , Yong Chen ·

YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2412.16615

Despite the recent advancement in Retrieval-Augmented Generation (RAG) systems, most retrieval methodologies are often developed for factual retrieval, which assumes query and positive documents are semantically similar. In this paper, we instead propose and study a more challenging type of retrieval task, called hidden rationale retrieval, in which query and document are not similar but can be inferred by reasoning chains, logic relationships, or empirical experiences. To address such problems, an instruction-tuned Large language model (LLM) with a cross-encoder architecture could be a reasonable choice. To further strengthen pioneering LLM-based retrievers, we design a special instruction that transforms the retrieval task into a generative task by prompting LLM to answer a binary-choice question. The model can be fine-tuned with direct preference optimization (DPO). The framework is also optimized for computational efficiency with no performance degradation. We name this retrieval framework by RaHoRe and verify its zero-shot and fine-tuned performance superiority on Emotional Support Conversation (ESC), compared with previous retrieval works. Our study suggests the potential to employ LLM as a foundation for a wider scope of retrieval tasks. Our codes, models, and datasets are available on https://github.com/flyfree5/LaHoRe.

Related Topics

Computer Science

Artificial Intelligence

Data Science

History

Archaeology

Concepts

Foundation (evidence) Computer science Natural language processing Artificial intelligence Data science History Archaeology

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2412.16615
PDF: https://arxiv.org/pdf/2412.16615
OA Status: green
Related Works: 10
OpenAlex ID: https://openalex.org/W4405766520

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4405766520

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2412.16615

Digital Object Identifier
Title: Large Language Model Can Be a Foundation for Hidden Rationale-Based Retrieval

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2024

Year of publication
Publication date: 2024-12-21

Full publication date if available
Authors: Luo Ji, Fulai Guo, Teng Chen, Qing Gu, Xiaoyu Wang, Ningyuan Xi, Yihong Wang, Peng Yu, Yue Zhao, Hongyang Lei, Zhonglin Jiang, Yong Chen

List of authors in order
Landing page: https://arxiv.org/abs/2412.16615

Publisher landing page
PDF URL: https://arxiv.org/pdf/2412.16615

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/2412.16615

Direct OA link when available
Concepts: Foundation (evidence), Computer science, Natural language processing, Artificial intelligence, Data science, History, Archaeology

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/W4405766520
doi	https://doi.org/10.48550/arxiv.2412.16615
ids.doi	https://doi.org/10.48550/arxiv.2412.16615
ids.openalex	https://openalex.org/W4405766520
fwci
type	preprint
title	Large Language Model Can Be a Foundation for Hidden Rationale-Based Retrieval
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10028
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9053999781608582
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	Topic Modeling
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C2780966255
concepts[0].level	2
concepts[0].score	0.8023055791854858
concepts[0].wikidata	https://www.wikidata.org/wiki/Q5474306
concepts[0].display_name	Foundation (evidence)
concepts[1].id	https://openalex.org/C41008148
concepts[1].level	0
concepts[1].score	0.5959909558296204
concepts[1].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[1].display_name	Computer science
concepts[2].id	https://openalex.org/C204321447
concepts[2].level	1
concepts[2].score	0.41173624992370605
concepts[2].wikidata	https://www.wikidata.org/wiki/Q30642
concepts[2].display_name	Natural language processing
concepts[3].id	https://openalex.org/C154945302
concepts[3].level	1
concepts[3].score	0.3889584243297577
concepts[3].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[3].display_name	Artificial intelligence
concepts[4].id	https://openalex.org/C2522767166
concepts[4].level	1
concepts[4].score	0.3314747214317322
concepts[4].wikidata	https://www.wikidata.org/wiki/Q2374463
concepts[4].display_name	Data science
concepts[5].id	https://openalex.org/C95457728
concepts[5].level	0
concepts[5].score	0.15657949447631836
concepts[5].wikidata	https://www.wikidata.org/wiki/Q309
concepts[5].display_name	History
concepts[6].id	https://openalex.org/C166957645
concepts[6].level	1
concepts[6].score	0.08710736036300659
concepts[6].wikidata	https://www.wikidata.org/wiki/Q23498
concepts[6].display_name	Archaeology
keywords[0].id	https://openalex.org/keywords/foundation
keywords[0].score	0.8023055791854858
keywords[0].display_name	Foundation (evidence)
keywords[1].id	https://openalex.org/keywords/computer-science
keywords[1].score	0.5959909558296204
keywords[1].display_name	Computer science
keywords[2].id	https://openalex.org/keywords/natural-language-processing
keywords[2].score	0.41173624992370605
keywords[2].display_name	Natural language processing
keywords[3].id	https://openalex.org/keywords/artificial-intelligence
keywords[3].score	0.3889584243297577
keywords[3].display_name	Artificial intelligence
keywords[4].id	https://openalex.org/keywords/data-science
keywords[4].score	0.3314747214317322
keywords[4].display_name	Data science
keywords[5].id	https://openalex.org/keywords/history
keywords[5].score	0.15657949447631836
keywords[5].display_name	History
keywords[6].id	https://openalex.org/keywords/archaeology
keywords[6].score	0.08710736036300659
keywords[6].display_name	Archaeology
language	en
locations[0].id	pmh:oai:arXiv.org:2412.16615
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/2412.16615
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/2412.16615
locations[1].id	doi:10.48550/arxiv.2412.16615
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.2412.16615
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5100842451
authorships[0].author.orcid
authorships[0].author.display_name	Luo Ji
authorships[0].author_position	first
authorships[0].raw_author_name	Ji, Luo
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5050912053
authorships[1].author.orcid	https://orcid.org/0000-0003-1474-8899
authorships[1].author.display_name	Fulai Guo
authorships[1].author_position	middle
authorships[1].raw_author_name	Guo, Feixiang
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5100637767
authorships[2].author.orcid	https://orcid.org/0000-0001-6428-7790
authorships[2].author.display_name	Teng Chen
authorships[2].author_position	middle
authorships[2].raw_author_name	Chen, Teng
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5066574703
authorships[3].author.orcid	https://orcid.org/0000-0002-5948-4690
authorships[3].author.display_name	Qing Gu
authorships[3].author_position	middle
authorships[3].raw_author_name	Gu, Qingqing
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5100357573
authorships[4].author.orcid	https://orcid.org/0000-0001-9411-214X
authorships[4].author.display_name	Xiaoyu Wang
authorships[4].author_position	middle
authorships[4].raw_author_name	Wang, Xiaoyu
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5004687734
authorships[5].author.orcid
authorships[5].author.display_name	Ningyuan Xi
authorships[5].author_position	middle
authorships[5].raw_author_name	Xi, Ningyuan
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5053640523
authorships[6].author.orcid	https://orcid.org/0000-0001-8407-061X
authorships[6].author.display_name	Yihong Wang
authorships[6].author_position	middle
authorships[6].raw_author_name	Wang, Yihong
authorships[6].is_corresponding	False
authorships[7].author.id	https://openalex.org/A5020189974
authorships[7].author.orcid	https://orcid.org/0000-0001-6310-514X
authorships[7].author.display_name	Peng Yu
authorships[7].author_position	middle
authorships[7].raw_author_name	Yu, Peng
authorships[7].is_corresponding	False
authorships[8].author.id	https://openalex.org/A5107951791
authorships[8].author.orcid
authorships[8].author.display_name	Yue Zhao
authorships[8].author_position	middle
authorships[8].raw_author_name	Zhao, Yue
authorships[8].is_corresponding	False
authorships[9].author.id	https://openalex.org/A5103246867
authorships[9].author.orcid	https://orcid.org/0009-0001-9705-4595
authorships[9].author.display_name	Hongyang Lei
authorships[9].author_position	middle
authorships[9].raw_author_name	Lei, Hongyang
authorships[9].is_corresponding	False
authorships[10].author.id	https://openalex.org/A5104274609
authorships[10].author.orcid
authorships[10].author.display_name	Zhonglin Jiang
authorships[10].author_position	middle
authorships[10].raw_author_name	Jiang, Zhonglin
authorships[10].is_corresponding	False
authorships[11].author.id	https://openalex.org/A5035121805
authorships[11].author.orcid	https://orcid.org/0000-0003-1691-9460
authorships[11].author.display_name	Yong Chen
authorships[11].author_position	last
authorships[11].raw_author_name	Chen, Yong
authorships[11].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/2412.16615
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	Large Language Model Can Be a Foundation for Hidden Rationale-Based Retrieval
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10028
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9053999781608582
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	Topic Modeling
related_works	https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2381393187, https://openalex.org/W2332779545, https://openalex.org/W2390279801, https://openalex.org/W4391913857, https://openalex.org/W2358668433, https://openalex.org/W2358060160, https://openalex.org/W3204019825
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2412.16615
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/2412.16615
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/2412.16615
primary_location.id	pmh:oai:arXiv.org:2412.16615
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/2412.16615
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/2412.16615
publication_date	2024-12-21
publication_year	2024
referenced_works_count	0
abstract_inverted_index.a	35, 77, 82, 93, 102, 110, 169, 172
abstract_inverted_index.In	27
abstract_inverted_index.To	66, 85
abstract_inverted_index.We	135
abstract_inverted_index.an	70
abstract_inverted_index.as	168
abstract_inverted_index.be	56, 81, 116
abstract_inverted_index.by	58, 105, 140
abstract_inverted_index.in	4, 46
abstract_inverted_index.is	125
abstract_inverted_index.no	132
abstract_inverted_index.of	39, 175
abstract_inverted_index.on	150, 185
abstract_inverted_index.or	63
abstract_inverted_index.to	108, 165
abstract_inverted_index.we	30, 91
abstract_inverted_index.LLM	107, 167
abstract_inverted_index.Our	160, 178
abstract_inverted_index.The	113, 123
abstract_inverted_index.and	21, 33, 49, 142, 146, 181
abstract_inverted_index.are	12, 24, 51, 183
abstract_inverted_index.but	54
abstract_inverted_index.can	55, 115
abstract_inverted_index.for	15, 128, 171
abstract_inverted_index.its	144
abstract_inverted_index.not	52
abstract_inverted_index.the	1, 98, 163
abstract_inverted_index.also	126
abstract_inverted_index.into	101
abstract_inverted_index.more	36
abstract_inverted_index.most	9
abstract_inverted_index.name	136
abstract_inverted_index.such	68
abstract_inverted_index.task	100, 104
abstract_inverted_index.that	96
abstract_inverted_index.this	28, 137
abstract_inverted_index.type	38
abstract_inverted_index.with	76, 118, 131, 156
abstract_inverted_index.(LLM)	75
abstract_inverted_index.(RAG)	7
abstract_inverted_index.Large	72
abstract_inverted_index.could	80
abstract_inverted_index.logic	61
abstract_inverted_index.model	74, 114
abstract_inverted_index.often	13
abstract_inverted_index.query	20, 48
abstract_inverted_index.scope	174
abstract_inverted_index.study	34, 161
abstract_inverted_index.task,	41
abstract_inverted_index.which	18, 47
abstract_inverted_index.wider	173
abstract_inverted_index.(DPO).	122
abstract_inverted_index.(ESC),	154
abstract_inverted_index.RaHoRe	141
abstract_inverted_index.answer	109
abstract_inverted_index.called	42
abstract_inverted_index.codes,	179
abstract_inverted_index.design	92
abstract_inverted_index.direct	119
abstract_inverted_index.employ	166
abstract_inverted_index.hidden	43
abstract_inverted_index.paper,	29
abstract_inverted_index.recent	2
abstract_inverted_index.tasks.	177
abstract_inverted_index.verify	143
abstract_inverted_index.works.	159
abstract_inverted_index.Despite	0
abstract_inverted_index.Support	152
abstract_inverted_index.address	67
abstract_inverted_index.assumes	19
abstract_inverted_index.chains,	60
abstract_inverted_index.choice.	84
abstract_inverted_index.factual	16
abstract_inverted_index.further	86
abstract_inverted_index.instead	31
abstract_inverted_index.models,	180
abstract_inverted_index.propose	32
abstract_inverted_index.similar	53
abstract_inverted_index.special	94
abstract_inverted_index.compared	155
abstract_inverted_index.datasets	182
abstract_inverted_index.document	50
abstract_inverted_index.inferred	57
abstract_inverted_index.language	73
abstract_inverted_index.positive	22
abstract_inverted_index.previous	157
abstract_inverted_index.similar.	26
abstract_inverted_index.suggests	162
abstract_inverted_index.systems,	8
abstract_inverted_index.Emotional	151
abstract_inverted_index.LLM-based	89
abstract_inverted_index.available	184
abstract_inverted_index.developed	14
abstract_inverted_index.documents	23
abstract_inverted_index.empirical	64
abstract_inverted_index.framework	124, 139
abstract_inverted_index.optimized	127
abstract_inverted_index.potential	164
abstract_inverted_index.problems,	69
abstract_inverted_index.prompting	106
abstract_inverted_index.question.	112
abstract_inverted_index.rationale	44
abstract_inverted_index.reasoning	59
abstract_inverted_index.retrieval	10, 40, 99, 138, 158, 176
abstract_inverted_index.zero-shot	145
abstract_inverted_index.Generation	6
abstract_inverted_index.efficiency	130
abstract_inverted_index.fine-tuned	117, 147
abstract_inverted_index.foundation	170
abstract_inverted_index.generative	103
abstract_inverted_index.pioneering	88
abstract_inverted_index.preference	120
abstract_inverted_index.reasonable	83
abstract_inverted_index.retrieval,	17, 45
abstract_inverted_index.strengthen	87
abstract_inverted_index.transforms	97
abstract_inverted_index.advancement	3
abstract_inverted_index.challenging	37
abstract_inverted_index.instruction	95
abstract_inverted_index.performance	133, 148
abstract_inverted_index.retrievers,	90
abstract_inverted_index.superiority	149
abstract_inverted_index.Conversation	153
abstract_inverted_index.architecture	79
abstract_inverted_index.degradation.	134
abstract_inverted_index.experiences.	65
abstract_inverted_index.optimization	121
abstract_inverted_index.semantically	25
abstract_inverted_index.binary-choice	111
abstract_inverted_index.computational	129
abstract_inverted_index.cross-encoder	78
abstract_inverted_index.methodologies	11
abstract_inverted_index.relationships,	62
abstract_inverted_index.instruction-tuned	71
abstract_inverted_index.Retrieval-Augmented	5
abstract_inverted_index.https://github.com/flyfree5/LaHoRe.	186
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	12
citation_normalized_percentile