COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation Article Swipe

PDF

Xinghong Liu , Yi Zhou , Tao Zhou , Chun-Mei Feng , Ling Shao ·

YOU? · · 2023 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2308.10450

Universal domain adaptation (UniDA) aims to address domain and category shifts across data sources. Recently, due to more stringent data restrictions, researchers have introduced source-free UniDA (SF-UniDA). SF-UniDA methods eliminate the need for direct access to source samples when performing adaptation to the target domain. However, existing SF-UniDA methods still require an extensive quantity of labeled source samples to train a source model, resulting in significant labeling costs. To tackle this issue, we present a novel plug-and-play classifier-oriented calibration (COCA) method. COCA, which exploits textual prototypes, is designed for the source models based on few-shot learning with vision-language models (VLMs). It endows the VLM-powered few-shot learners, which are built for closed-set classification, with the unknown-aware ability to distinguish common and unknown classes in the SF-UniDA scenario. Crucially, COCA is a new paradigm to tackle SF-UniDA challenges based on VLMs, which focuses on classifier instead of image encoder optimization. Experiments show that COCA outperforms state-of-the-art UniDA and SF-UniDA models.

Related Topics

Computer Science

Artificial Intelligence

Concepts

Classifier (UML) Domain adaptation Computer science Geology Artificial intelligence

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2308.10450
PDF: https://arxiv.org/pdf/2308.10450
OA Status: green
Related Works: 10
OpenAlex ID: https://openalex.org/W4386081439

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4386081439

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2308.10450

Digital Object Identifier
Title: COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2023

Year of publication
Publication date: 2023-08-21

Full publication date if available
Authors: Xinghong Liu, Yi Zhou, Tao Zhou, Chun-Mei Feng, Ling Shao

List of authors in order
Landing page: https://arxiv.org/abs/2308.10450

Publisher landing page
PDF URL: https://arxiv.org/pdf/2308.10450

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/2308.10450

Direct OA link when available
Concepts: Classifier (UML), Domain adaptation, Computer science, Geology, 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/W4386081439
doi	https://doi.org/10.48550/arxiv.2308.10450
ids.doi	https://doi.org/10.48550/arxiv.2308.10450
ids.openalex	https://openalex.org/W4386081439
fwci
type	preprint
title	COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T11307
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9876999855041504
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	Domain Adaptation and Few-Shot Learning
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C95623464
concepts[0].level	2
concepts[0].score	0.46168914437294006
concepts[0].wikidata	https://www.wikidata.org/wiki/Q1096149
concepts[0].display_name	Classifier (UML)
concepts[1].id	https://openalex.org/C2776434776
concepts[1].level	3
concepts[1].score	0.4428936839103699
concepts[1].wikidata	https://www.wikidata.org/wiki/Q19246213
concepts[1].display_name	Domain adaptation
concepts[2].id	https://openalex.org/C41008148
concepts[2].level	0
concepts[2].score	0.43494054675102234
concepts[2].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[2].display_name	Computer science
concepts[3].id	https://openalex.org/C127313418
concepts[3].level	0
concepts[3].score	0.3737720251083374
concepts[3].wikidata	https://www.wikidata.org/wiki/Q1069
concepts[3].display_name	Geology
concepts[4].id	https://openalex.org/C154945302
concepts[4].level	1
concepts[4].score	0.3416213393211365
concepts[4].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[4].display_name	Artificial intelligence
keywords[0].id	https://openalex.org/keywords/classifier
keywords[0].score	0.46168914437294006
keywords[0].display_name	Classifier (UML)
keywords[1].id	https://openalex.org/keywords/domain-adaptation
keywords[1].score	0.4428936839103699
keywords[1].display_name	Domain adaptation
keywords[2].id	https://openalex.org/keywords/computer-science
keywords[2].score	0.43494054675102234
keywords[2].display_name	Computer science
keywords[3].id	https://openalex.org/keywords/geology
keywords[3].score	0.3737720251083374
keywords[3].display_name	Geology
keywords[4].id	https://openalex.org/keywords/artificial-intelligence
keywords[4].score	0.3416213393211365
keywords[4].display_name	Artificial intelligence
language	en
locations[0].id	pmh:oai:arXiv.org:2308.10450
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/2308.10450
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/2308.10450
locations[1].id	doi:10.48550/arxiv.2308.10450
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.2308.10450
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5047555766
authorships[0].author.orcid	https://orcid.org/0000-0002-1627-7667
authorships[0].author.display_name	Xinghong Liu
authorships[0].author_position	first
authorships[0].raw_author_name	Liu, Xinghong
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5100753053
authorships[1].author.orcid	https://orcid.org/0000-0003-0020-847X
authorships[1].author.display_name	Yi Zhou
authorships[1].author_position	middle
authorships[1].raw_author_name	Zhou, Yi
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5067280360
authorships[2].author.orcid	https://orcid.org/0000-0002-8145-712X
authorships[2].author.display_name	Tao Zhou
authorships[2].author_position	middle
authorships[2].raw_author_name	Zhou, Tao
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5049444898
authorships[3].author.orcid	https://orcid.org/0000-0002-3025-8964
authorships[3].author.display_name	Chun-Mei Feng
authorships[3].author_position	middle
authorships[3].raw_author_name	Feng, Chun-Mei
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5082634513
authorships[4].author.orcid	https://orcid.org/0000-0002-8264-6117
authorships[4].author.display_name	Ling Shao
authorships[4].author_position	last
authorships[4].raw_author_name	Shao, Ling
authorships[4].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/2308.10450
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2023-08-23T00:00:00
display_name	COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T11307
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9876999855041504
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	Domain Adaptation and Few-Shot Learning
related_works	https://openalex.org/W3080655457, https://openalex.org/W2145868540, https://openalex.org/W3166286441, https://openalex.org/W3214142563, https://openalex.org/W3136267388, https://openalex.org/W3186065094, https://openalex.org/W4287263085, https://openalex.org/W3093803318, https://openalex.org/W3204418343, https://openalex.org/W4390401377
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2308.10450
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/2308.10450
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/2308.10450
primary_location.id	pmh:oai:arXiv.org:2308.10450
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/2308.10450
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/2308.10450
publication_date	2023-08-21
publication_year	2023
referenced_works_count	0
abstract_inverted_index.a	60, 74, 129
abstract_inverted_index.It	100
abstract_inverted_index.To	68
abstract_inverted_index.an	51
abstract_inverted_index.in	64, 122
abstract_inverted_index.is	86, 128
abstract_inverted_index.of	54, 144
abstract_inverted_index.on	93, 137, 141
abstract_inverted_index.to	5, 16, 35, 41, 58, 116, 132
abstract_inverted_index.we	72
abstract_inverted_index.and	8, 119, 155
abstract_inverted_index.are	107
abstract_inverted_index.due	15
abstract_inverted_index.for	32, 88, 109
abstract_inverted_index.new	130
abstract_inverted_index.the	30, 42, 89, 102, 113, 123
abstract_inverted_index.COCA	127, 151
abstract_inverted_index.aims	4
abstract_inverted_index.data	12, 19
abstract_inverted_index.have	22
abstract_inverted_index.more	17
abstract_inverted_index.need	31
abstract_inverted_index.show	149
abstract_inverted_index.that	150
abstract_inverted_index.this	70
abstract_inverted_index.when	38
abstract_inverted_index.with	96, 112
abstract_inverted_index.COCA,	81
abstract_inverted_index.UniDA	25, 154
abstract_inverted_index.VLMs,	138
abstract_inverted_index.based	92, 136
abstract_inverted_index.built	108
abstract_inverted_index.image	145
abstract_inverted_index.novel	75
abstract_inverted_index.still	49
abstract_inverted_index.train	59
abstract_inverted_index.which	82, 106, 139
abstract_inverted_index.(COCA)	79
abstract_inverted_index.access	34
abstract_inverted_index.across	11
abstract_inverted_index.common	118
abstract_inverted_index.costs.	67
abstract_inverted_index.direct	33
abstract_inverted_index.domain	1, 7
abstract_inverted_index.endows	101
abstract_inverted_index.issue,	71
abstract_inverted_index.model,	62
abstract_inverted_index.models	91, 98
abstract_inverted_index.shifts	10
abstract_inverted_index.source	36, 56, 61, 90
abstract_inverted_index.tackle	69, 133
abstract_inverted_index.target	43
abstract_inverted_index.(UniDA)	3
abstract_inverted_index.(VLMs).	99
abstract_inverted_index.ability	115
abstract_inverted_index.address	6
abstract_inverted_index.classes	121
abstract_inverted_index.domain.	44
abstract_inverted_index.encoder	146
abstract_inverted_index.focuses	140
abstract_inverted_index.instead	143
abstract_inverted_index.labeled	55
abstract_inverted_index.method.	80
abstract_inverted_index.methods	28, 48
abstract_inverted_index.models.	157
abstract_inverted_index.present	73
abstract_inverted_index.require	50
abstract_inverted_index.samples	37, 57
abstract_inverted_index.textual	84
abstract_inverted_index.unknown	120
abstract_inverted_index.However,	45
abstract_inverted_index.SF-UniDA	27, 47, 124, 134, 156
abstract_inverted_index.category	9
abstract_inverted_index.designed	87
abstract_inverted_index.existing	46
abstract_inverted_index.exploits	83
abstract_inverted_index.few-shot	94, 104
abstract_inverted_index.labeling	66
abstract_inverted_index.learning	95
abstract_inverted_index.paradigm	131
abstract_inverted_index.quantity	53
abstract_inverted_index.sources.	13
abstract_inverted_index.Recently,	14
abstract_inverted_index.Universal	0
abstract_inverted_index.eliminate	29
abstract_inverted_index.extensive	52
abstract_inverted_index.learners,	105
abstract_inverted_index.resulting	63
abstract_inverted_index.scenario.	125
abstract_inverted_index.stringent	18
abstract_inverted_index.Crucially,	126
abstract_inverted_index.adaptation	2, 40
abstract_inverted_index.challenges	135
abstract_inverted_index.classifier	142
abstract_inverted_index.closed-set	110
abstract_inverted_index.introduced	23
abstract_inverted_index.performing	39
abstract_inverted_index.(SF-UniDA).	26
abstract_inverted_index.Experiments	148
abstract_inverted_index.VLM-powered	103
abstract_inverted_index.calibration	78
abstract_inverted_index.distinguish	117
abstract_inverted_index.outperforms	152
abstract_inverted_index.prototypes,	85
abstract_inverted_index.researchers	21
abstract_inverted_index.significant	65
abstract_inverted_index.source-free	24
abstract_inverted_index.optimization.	147
abstract_inverted_index.plug-and-play	76
abstract_inverted_index.restrictions,	20
abstract_inverted_index.unknown-aware	114
abstract_inverted_index.classification,	111
abstract_inverted_index.vision-language	97
abstract_inverted_index.state-of-the-art	153
abstract_inverted_index.classifier-oriented	77
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	5
citation_normalized_percentile