Enhancing Clean Label Backdoor Attack with Two-phase Specific Triggers Article Swipe

PDF

Nan Luo , Yuanzhang Li , Yajie Wang , Shangbo Wu , Yu‐an Tan , Quanxin Zhang ·

YOU? · · 2022 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2206.04881

Backdoor attacks threaten Deep Neural Networks (DNNs). Towards stealthiness, researchers propose clean-label backdoor attacks, which require the adversaries not to alter the labels of the poisoned training datasets. Clean-label settings make the attack more stealthy due to the correct image-label pairs, but some problems still exist: first, traditional methods for poisoning training data are ineffective; second, traditional triggers are not stealthy which are still perceptible. To solve these problems, we propose a two-phase and image-specific triggers generation method to enhance clean-label backdoor attacks. Our methods are (1) powerful: our triggers can both promote the two phases (i.e., the backdoor implantation and activation phase) in backdoor attacks simultaneously; (2) stealthy: our triggers are generated from each image. They are image-specific instead of fixed triggers. Extensive experiments demonstrate that our approach can achieve a fantastic attack success rate~(98.98%) with low poisoning rate~(5%), high stealthiness under many evaluation metrics and is resistant to backdoor defense methods.

Related Topics

Computer Science

Computer Security

Phase (Matter)

Artificial Intelligence

Physics

Quantum Mechanics

Concepts

Backdoor Computer science Computer security Image (mathematics) Phase (matter) Artificial intelligence Physics Quantum mechanics

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2206.04881
PDF: https://arxiv.org/pdf/2206.04881
OA Status: green
Cited By: 5
Related Works: 10
OpenAlex ID: https://openalex.org/W4282813371

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4282813371

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2206.04881

Digital Object Identifier
Title: Enhancing Clean Label Backdoor Attack with Two-phase Specific Triggers

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2022

Year of publication
Publication date: 2022-06-10

Full publication date if available
Authors: Nan Luo, Yuanzhang Li, Yajie Wang, Shangbo Wu, Yu‐an Tan, Quanxin Zhang

List of authors in order
Landing page: https://arxiv.org/abs/2206.04881

Publisher landing page
PDF URL: https://arxiv.org/pdf/2206.04881

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/2206.04881

Direct OA link when available
Concepts: Backdoor, Computer science, Computer security, Image (mathematics), Phase (matter), Artificial intelligence, Physics, Quantum mechanics

Top concepts (fields/topics) attached by OpenAlex
Cited by: 5

Total citation count in OpenAlex
Citations by year (recent): 2024: 3, 2023: 2

Per-year citation counts (last 5 years)
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4282813371
doi	https://doi.org/10.48550/arxiv.2206.04881
ids.doi	https://doi.org/10.48550/arxiv.2206.04881
ids.openalex	https://openalex.org/W4282813371
fwci
type	preprint
title	Enhancing Clean Label Backdoor Attack with Two-phase Specific Triggers
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T11689
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9998000264167786
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	Adversarial Robustness in Machine Learning
topics[1].id	https://openalex.org/T11512
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.9815999865531921
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1702
topics[1].subfield.display_name	Artificial Intelligence
topics[1].display_name	Anomaly Detection Techniques and Applications
topics[2].id	https://openalex.org/T10036
topics[2].field.id	https://openalex.org/fields/17
topics[2].field.display_name	Computer Science
topics[2].score	0.9722999930381775
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	Advanced Neural Network Applications
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C2781045450
concepts[0].level	2
concepts[0].score	0.9993759393692017
concepts[0].wikidata	https://www.wikidata.org/wiki/Q254569
concepts[0].display_name	Backdoor
concepts[1].id	https://openalex.org/C41008148
concepts[1].level	0
concepts[1].score	0.5978050827980042
concepts[1].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[1].display_name	Computer science
concepts[2].id	https://openalex.org/C38652104
concepts[2].level	1
concepts[2].score	0.5734850168228149
concepts[2].wikidata	https://www.wikidata.org/wiki/Q3510521
concepts[2].display_name	Computer security
concepts[3].id	https://openalex.org/C115961682
concepts[3].level	2
concepts[3].score	0.5445570945739746
concepts[3].wikidata	https://www.wikidata.org/wiki/Q860623
concepts[3].display_name	Image (mathematics)
concepts[4].id	https://openalex.org/C44280652
concepts[4].level	2
concepts[4].score	0.4147685170173645
concepts[4].wikidata	https://www.wikidata.org/wiki/Q104837
concepts[4].display_name	Phase (matter)
concepts[5].id	https://openalex.org/C154945302
concepts[5].level	1
concepts[5].score	0.31894177198410034
concepts[5].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[5].display_name	Artificial intelligence
concepts[6].id	https://openalex.org/C121332964
concepts[6].level	0
concepts[6].score	0.11376067996025085
concepts[6].wikidata	https://www.wikidata.org/wiki/Q413
concepts[6].display_name	Physics
concepts[7].id	https://openalex.org/C62520636
concepts[7].level	1
concepts[7].score	0.0
concepts[7].wikidata	https://www.wikidata.org/wiki/Q944
concepts[7].display_name	Quantum mechanics
keywords[0].id	https://openalex.org/keywords/backdoor
keywords[0].score	0.9993759393692017
keywords[0].display_name	Backdoor
keywords[1].id	https://openalex.org/keywords/computer-science
keywords[1].score	0.5978050827980042
keywords[1].display_name	Computer science
keywords[2].id	https://openalex.org/keywords/computer-security
keywords[2].score	0.5734850168228149
keywords[2].display_name	Computer security
keywords[3].id	https://openalex.org/keywords/image
keywords[3].score	0.5445570945739746
keywords[3].display_name	Image (mathematics)
keywords[4].id	https://openalex.org/keywords/phase
keywords[4].score	0.4147685170173645
keywords[4].display_name	Phase (matter)
keywords[5].id	https://openalex.org/keywords/artificial-intelligence
keywords[5].score	0.31894177198410034
keywords[5].display_name	Artificial intelligence
keywords[6].id	https://openalex.org/keywords/physics
keywords[6].score	0.11376067996025085
keywords[6].display_name	Physics
language	en
locations[0].id	pmh:oai:arXiv.org:2206.04881
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/2206.04881
locations[0].version	submittedVersion
locations[0].raw_type
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/2206.04881
locations[1].id	doi:10.48550/arxiv.2206.04881
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.2206.04881
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5064398143
authorships[0].author.orcid	https://orcid.org/0000-0001-7980-6979
authorships[0].author.display_name	Nan Luo
authorships[0].author_position	first
authorships[0].raw_author_name	Luo, Nan
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5026287089
authorships[1].author.orcid	https://orcid.org/0000-0002-1931-366X
authorships[1].author.display_name	Yuanzhang Li
authorships[1].author_position	middle
authorships[1].raw_author_name	Li, Yuanzhang
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5100455413
authorships[2].author.orcid	https://orcid.org/0000-0002-0962-4464
authorships[2].author.display_name	Yajie Wang
authorships[2].author_position	middle
authorships[2].raw_author_name	Wang, Yajie
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5090304311
authorships[3].author.orcid	https://orcid.org/0000-0002-0737-7420
authorships[3].author.display_name	Shangbo Wu
authorships[3].author_position	middle
authorships[3].raw_author_name	Wu, Shangbo
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5003198271
authorships[4].author.orcid	https://orcid.org/0000-0001-6404-8853
authorships[4].author.display_name	Yu‐an Tan
authorships[4].author_position	middle
authorships[4].raw_author_name	Tan, Yu-an
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5017675981
authorships[5].author.orcid	https://orcid.org/0000-0002-5094-7388
authorships[5].author.display_name	Quanxin Zhang
authorships[5].author_position	last
authorships[5].raw_author_name	Zhang, Quanxin
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/2206.04881
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	Enhancing Clean Label Backdoor Attack with Two-phase Specific Triggers
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T11689
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9998000264167786
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	Adversarial Robustness in Machine Learning
related_works	https://openalex.org/W4320031223, https://openalex.org/W4200629851, https://openalex.org/W4281902577, https://openalex.org/W4309417370, https://openalex.org/W4292107232, https://openalex.org/W3009072493, https://openalex.org/W4386080799, https://openalex.org/W3140988292, https://openalex.org/W4317672133, https://openalex.org/W4386185023
cited_by_count	5
counts_by_year[0].year	2024
counts_by_year[0].cited_by_count	3
counts_by_year[1].year	2023
counts_by_year[1].cited_by_count	2
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2206.04881
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/2206.04881
best_oa_location.version	submittedVersion
best_oa_location.raw_type
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/2206.04881
primary_location.id	pmh:oai:arXiv.org:2206.04881
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/2206.04881
primary_location.version	submittedVersion
primary_location.raw_type
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/2206.04881
publication_date	2022-06-10
publication_year	2022
referenced_works_count	0
abstract_inverted_index.a	71, 131
abstract_inverted_index.To	65
abstract_inverted_index.in	103
abstract_inverted_index.is	147
abstract_inverted_index.of	23, 120
abstract_inverted_index.to	19, 36, 78, 149
abstract_inverted_index.we	69
abstract_inverted_index.(1)	86
abstract_inverted_index.(2)	107
abstract_inverted_index.Our	83
abstract_inverted_index.and	73, 100, 146
abstract_inverted_index.are	53, 58, 62, 85, 111, 117
abstract_inverted_index.but	41
abstract_inverted_index.can	90, 129
abstract_inverted_index.due	35
abstract_inverted_index.for	49
abstract_inverted_index.low	137
abstract_inverted_index.not	18, 59
abstract_inverted_index.our	88, 109, 127
abstract_inverted_index.the	16, 21, 24, 31, 37, 93, 97
abstract_inverted_index.two	94
abstract_inverted_index.Deep	3
abstract_inverted_index.They	116
abstract_inverted_index.both	91
abstract_inverted_index.data	52
abstract_inverted_index.each	114
abstract_inverted_index.from	113
abstract_inverted_index.high	140
abstract_inverted_index.make	30
abstract_inverted_index.many	143
abstract_inverted_index.more	33
abstract_inverted_index.some	42
abstract_inverted_index.that	126
abstract_inverted_index.with	136
abstract_inverted_index.alter	20
abstract_inverted_index.fixed	121
abstract_inverted_index.solve	66
abstract_inverted_index.still	44, 63
abstract_inverted_index.these	67
abstract_inverted_index.under	142
abstract_inverted_index.which	14, 61
abstract_inverted_index.(i.e.,	96
abstract_inverted_index.Neural	4
abstract_inverted_index.attack	32, 133
abstract_inverted_index.exist:	45
abstract_inverted_index.first,	46
abstract_inverted_index.image.	115
abstract_inverted_index.labels	22
abstract_inverted_index.method	77
abstract_inverted_index.pairs,	40
abstract_inverted_index.phase)	102
abstract_inverted_index.phases	95
abstract_inverted_index.(DNNs).	6
abstract_inverted_index.Towards	7
abstract_inverted_index.achieve	130
abstract_inverted_index.attacks	1, 105
abstract_inverted_index.correct	38
abstract_inverted_index.defense	151
abstract_inverted_index.enhance	79
abstract_inverted_index.instead	119
abstract_inverted_index.methods	48, 84
abstract_inverted_index.metrics	145
abstract_inverted_index.promote	92
abstract_inverted_index.propose	10, 70
abstract_inverted_index.require	15
abstract_inverted_index.second,	55
abstract_inverted_index.success	134
abstract_inverted_index.Backdoor	0
abstract_inverted_index.Networks	5
abstract_inverted_index.approach	128
abstract_inverted_index.attacks,	13
abstract_inverted_index.attacks.	82
abstract_inverted_index.backdoor	12, 81, 98, 104, 150
abstract_inverted_index.methods.	152
abstract_inverted_index.poisoned	25
abstract_inverted_index.problems	43
abstract_inverted_index.settings	29
abstract_inverted_index.stealthy	34, 60
abstract_inverted_index.threaten	2
abstract_inverted_index.training	26, 51
abstract_inverted_index.triggers	57, 75, 89, 110
abstract_inverted_index.Extensive	123
abstract_inverted_index.datasets.	27
abstract_inverted_index.fantastic	132
abstract_inverted_index.generated	112
abstract_inverted_index.poisoning	50, 138
abstract_inverted_index.powerful:	87
abstract_inverted_index.problems,	68
abstract_inverted_index.resistant	148
abstract_inverted_index.stealthy:	108
abstract_inverted_index.triggers.	122
abstract_inverted_index.two-phase	72
abstract_inverted_index.activation	101
abstract_inverted_index.evaluation	144
abstract_inverted_index.generation	76
abstract_inverted_index.rate~(5%),	139
abstract_inverted_index.Clean-label	28
abstract_inverted_index.adversaries	17
abstract_inverted_index.clean-label	11, 80
abstract_inverted_index.demonstrate	125
abstract_inverted_index.experiments	124
abstract_inverted_index.image-label	39
abstract_inverted_index.researchers	9
abstract_inverted_index.traditional	47, 56
abstract_inverted_index.implantation	99
abstract_inverted_index.ineffective;	54
abstract_inverted_index.perceptible.	64
abstract_inverted_index.stealthiness	141
abstract_inverted_index.rate~(98.98%)	135
abstract_inverted_index.stealthiness,	8
abstract_inverted_index.image-specific	74, 118
abstract_inverted_index.simultaneously;	106
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	6
citation_normalized_percentile