Language Guided Adversarial Purification Article Swipe
Himanshu Singh
,
A V Subramanyam
·
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2309.10348
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2309.10348
Adversarial purification using generative models demonstrates strong adversarial defense performance. These methods are classifier and attack-agnostic, making them versatile but often computationally intensive. Recent strides in diffusion and score networks have improved image generation and, by extension, adversarial purification. Another highly efficient class of adversarial defense methods known as adversarial training requires specific knowledge of attack vectors, forcing them to be trained extensively on adversarial examples. To overcome these limitations, we introduce a new framework, namely Language Guided Adversarial Purification (LGAP), utilizing pre-trained diffusion models and caption generators to defend against adversarial attacks. Given an input image, our method first generates a caption, which is then used to guide the adversarial purification process through a diffusion network. Our approach has been evaluated against strong adversarial attacks, proving its effectiveness in enhancing adversarial robustness. Our results indicate that LGAP outperforms most existing adversarial defense techniques without requiring specialized network training. This underscores the generalizability of models trained on large datasets, highlighting a promising direction for further research.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2309.10348
- https://arxiv.org/pdf/2309.10348
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4386907383
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4386907383Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2309.10348Digital Object Identifier
- Title
-
Language Guided Adversarial PurificationWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2023Year of publication
- Publication date
-
2023-09-19Full publication date if available
- Authors
-
Himanshu Singh, A V SubramanyamList of authors in order
- Landing page
-
https://arxiv.org/abs/2309.10348Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2309.10348Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2309.10348Direct OA link when available
- Concepts
-
Adversarial system, Computer science, Generalizability theory, Artificial intelligence, Robustness (evolution), Generative adversarial network, Machine learning, Classifier (UML), Image translation, Generative grammar, Image (mathematics), Mathematics, Chemistry, Statistics, Gene, BiochemistryTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4386907383 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2309.10348 |
| ids.doi | https://doi.org/10.48550/arxiv.2309.10348 |
| ids.openalex | https://openalex.org/W4386907383 |
| fwci | |
| type | preprint |
| title | Language Guided Adversarial Purification |
| 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 | 0.9937999844551086 |
| 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/T10775 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9821000099182129 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1707 |
| topics[1].subfield.display_name | Computer Vision and Pattern Recognition |
| topics[1].display_name | Generative Adversarial Networks and Image Synthesis |
| topics[2].id | https://openalex.org/T11914 |
| topics[2].field.id | https://openalex.org/fields/27 |
| topics[2].field.display_name | Medicine |
| topics[2].score | 0.9085000157356262 |
| topics[2].domain.id | https://openalex.org/domains/4 |
| topics[2].domain.display_name | Health Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/2730 |
| topics[2].subfield.display_name | Oncology |
| topics[2].display_name | PARP inhibition in cancer therapy |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C37736160 |
| concepts[0].level | 2 |
| concepts[0].score | 0.9757913947105408 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q1801315 |
| concepts[0].display_name | Adversarial system |
| concepts[1].id | https://openalex.org/C41008148 |
| concepts[1].level | 0 |
| concepts[1].score | 0.7541307210922241 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[1].display_name | Computer science |
| concepts[2].id | https://openalex.org/C27158222 |
| concepts[2].level | 2 |
| concepts[2].score | 0.6366600394248962 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q5532422 |
| concepts[2].display_name | Generalizability theory |
| concepts[3].id | https://openalex.org/C154945302 |
| concepts[3].level | 1 |
| concepts[3].score | 0.6173285841941833 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[3].display_name | Artificial intelligence |
| concepts[4].id | https://openalex.org/C63479239 |
| concepts[4].level | 3 |
| concepts[4].score | 0.5044354200363159 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q7353546 |
| concepts[4].display_name | Robustness (evolution) |
| concepts[5].id | https://openalex.org/C2988773926 |
| concepts[5].level | 3 |
| concepts[5].score | 0.5021188259124756 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q25104379 |
| concepts[5].display_name | Generative adversarial network |
| concepts[6].id | https://openalex.org/C119857082 |
| concepts[6].level | 1 |
| concepts[6].score | 0.4938700795173645 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[6].display_name | Machine learning |
| concepts[7].id | https://openalex.org/C95623464 |
| concepts[7].level | 2 |
| concepts[7].score | 0.49091750383377075 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q1096149 |
| concepts[7].display_name | Classifier (UML) |
| concepts[8].id | https://openalex.org/C2779757391 |
| concepts[8].level | 3 |
| concepts[8].score | 0.4511064291000366 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q6002292 |
| concepts[8].display_name | Image translation |
| concepts[9].id | https://openalex.org/C39890363 |
| concepts[9].level | 2 |
| concepts[9].score | 0.41425979137420654 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q36108 |
| concepts[9].display_name | Generative grammar |
| concepts[10].id | https://openalex.org/C115961682 |
| concepts[10].level | 2 |
| concepts[10].score | 0.3549519181251526 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q860623 |
| concepts[10].display_name | Image (mathematics) |
| concepts[11].id | https://openalex.org/C33923547 |
| concepts[11].level | 0 |
| concepts[11].score | 0.1132047176361084 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[11].display_name | Mathematics |
| concepts[12].id | https://openalex.org/C185592680 |
| concepts[12].level | 0 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q2329 |
| concepts[12].display_name | Chemistry |
| concepts[13].id | https://openalex.org/C105795698 |
| concepts[13].level | 1 |
| concepts[13].score | 0.0 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q12483 |
| concepts[13].display_name | Statistics |
| concepts[14].id | https://openalex.org/C104317684 |
| concepts[14].level | 2 |
| concepts[14].score | 0.0 |
| concepts[14].wikidata | https://www.wikidata.org/wiki/Q7187 |
| concepts[14].display_name | Gene |
| concepts[15].id | https://openalex.org/C55493867 |
| concepts[15].level | 1 |
| concepts[15].score | 0.0 |
| concepts[15].wikidata | https://www.wikidata.org/wiki/Q7094 |
| concepts[15].display_name | Biochemistry |
| keywords[0].id | https://openalex.org/keywords/adversarial-system |
| keywords[0].score | 0.9757913947105408 |
| keywords[0].display_name | Adversarial system |
| keywords[1].id | https://openalex.org/keywords/computer-science |
| keywords[1].score | 0.7541307210922241 |
| keywords[1].display_name | Computer science |
| keywords[2].id | https://openalex.org/keywords/generalizability-theory |
| keywords[2].score | 0.6366600394248962 |
| keywords[2].display_name | Generalizability theory |
| keywords[3].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[3].score | 0.6173285841941833 |
| keywords[3].display_name | Artificial intelligence |
| keywords[4].id | https://openalex.org/keywords/robustness |
| keywords[4].score | 0.5044354200363159 |
| keywords[4].display_name | Robustness (evolution) |
| keywords[5].id | https://openalex.org/keywords/generative-adversarial-network |
| keywords[5].score | 0.5021188259124756 |
| keywords[5].display_name | Generative adversarial network |
| keywords[6].id | https://openalex.org/keywords/machine-learning |
| keywords[6].score | 0.4938700795173645 |
| keywords[6].display_name | Machine learning |
| keywords[7].id | https://openalex.org/keywords/classifier |
| keywords[7].score | 0.49091750383377075 |
| keywords[7].display_name | Classifier (UML) |
| keywords[8].id | https://openalex.org/keywords/image-translation |
| keywords[8].score | 0.4511064291000366 |
| keywords[8].display_name | Image translation |
| keywords[9].id | https://openalex.org/keywords/generative-grammar |
| keywords[9].score | 0.41425979137420654 |
| keywords[9].display_name | Generative grammar |
| keywords[10].id | https://openalex.org/keywords/image |
| keywords[10].score | 0.3549519181251526 |
| keywords[10].display_name | Image (mathematics) |
| keywords[11].id | https://openalex.org/keywords/mathematics |
| keywords[11].score | 0.1132047176361084 |
| keywords[11].display_name | Mathematics |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2309.10348 |
| 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 | cc-by |
| locations[0].pdf_url | https://arxiv.org/pdf/2309.10348 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | https://openalex.org/licenses/cc-by |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2309.10348 |
| locations[1].id | doi:10.48550/arxiv.2309.10348 |
| 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 | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| 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.2309.10348 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5100372937 |
| authorships[0].author.orcid | https://orcid.org/0009-0000-8946-3674 |
| authorships[0].author.display_name | Himanshu Singh |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Singh, Himanshu |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5085785393 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-8873-4644 |
| authorships[1].author.display_name | A V Subramanyam |
| authorships[1].author_position | last |
| authorships[1].raw_author_name | Subramanyam, A V |
| authorships[1].is_corresponding | False |
| has_content.pdf | True |
| has_content.grobid_xml | True |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2309.10348 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Language Guided Adversarial Purification |
| has_fulltext | True |
| 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 | 0.9937999844551086 |
| 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/W4312352990, https://openalex.org/W2888032422, https://openalex.org/W3156763702, https://openalex.org/W4385421777, https://openalex.org/W4377980832, https://openalex.org/W2905311601, https://openalex.org/W2897769091, https://openalex.org/W4297411772, https://openalex.org/W2845413374, https://openalex.org/W4226298148 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2309.10348 |
| 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 | cc-by |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2309.10348 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | https://openalex.org/licenses/cc-by |
| 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/2309.10348 |
| primary_location.id | pmh:oai:arXiv.org:2309.10348 |
| 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 | cc-by |
| primary_location.pdf_url | https://arxiv.org/pdf/2309.10348 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | https://openalex.org/licenses/cc-by |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2309.10348 |
| publication_date | 2023-09-19 |
| publication_year | 2023 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 72, 101, 114, 160 |
| abstract_inverted_index.To | 66 |
| abstract_inverted_index.an | 94 |
| abstract_inverted_index.as | 48 |
| abstract_inverted_index.be | 60 |
| abstract_inverted_index.by | 35 |
| abstract_inverted_index.in | 25, 129 |
| abstract_inverted_index.is | 104 |
| abstract_inverted_index.of | 43, 54, 153 |
| abstract_inverted_index.on | 63, 156 |
| abstract_inverted_index.to | 59, 88, 107 |
| abstract_inverted_index.we | 70 |
| abstract_inverted_index.Our | 117, 133 |
| abstract_inverted_index.and | 14, 27, 85 |
| abstract_inverted_index.are | 12 |
| abstract_inverted_index.but | 19 |
| abstract_inverted_index.for | 163 |
| abstract_inverted_index.has | 119 |
| abstract_inverted_index.its | 127 |
| abstract_inverted_index.new | 73 |
| abstract_inverted_index.our | 97 |
| abstract_inverted_index.the | 109, 151 |
| abstract_inverted_index.LGAP | 137 |
| abstract_inverted_index.This | 149 |
| abstract_inverted_index.and, | 34 |
| abstract_inverted_index.been | 120 |
| abstract_inverted_index.have | 30 |
| abstract_inverted_index.most | 139 |
| abstract_inverted_index.that | 136 |
| abstract_inverted_index.them | 17, 58 |
| abstract_inverted_index.then | 105 |
| abstract_inverted_index.used | 106 |
| abstract_inverted_index.Given | 93 |
| abstract_inverted_index.These | 10 |
| abstract_inverted_index.class | 42 |
| abstract_inverted_index.first | 99 |
| abstract_inverted_index.guide | 108 |
| abstract_inverted_index.image | 32 |
| abstract_inverted_index.input | 95 |
| abstract_inverted_index.known | 47 |
| abstract_inverted_index.large | 157 |
| abstract_inverted_index.often | 20 |
| abstract_inverted_index.score | 28 |
| abstract_inverted_index.these | 68 |
| abstract_inverted_index.using | 2 |
| abstract_inverted_index.which | 103 |
| abstract_inverted_index.Guided | 77 |
| abstract_inverted_index.Recent | 23 |
| abstract_inverted_index.attack | 55 |
| abstract_inverted_index.defend | 89 |
| abstract_inverted_index.highly | 40 |
| abstract_inverted_index.image, | 96 |
| abstract_inverted_index.making | 16 |
| abstract_inverted_index.method | 98 |
| abstract_inverted_index.models | 4, 84, 154 |
| abstract_inverted_index.namely | 75 |
| abstract_inverted_index.strong | 6, 123 |
| abstract_inverted_index.(LGAP), | 80 |
| abstract_inverted_index.Another | 39 |
| abstract_inverted_index.against | 90, 122 |
| abstract_inverted_index.caption | 86 |
| abstract_inverted_index.defense | 8, 45, 142 |
| abstract_inverted_index.forcing | 57 |
| abstract_inverted_index.further | 164 |
| abstract_inverted_index.methods | 11, 46 |
| abstract_inverted_index.network | 147 |
| abstract_inverted_index.process | 112 |
| abstract_inverted_index.proving | 126 |
| abstract_inverted_index.results | 134 |
| abstract_inverted_index.strides | 24 |
| abstract_inverted_index.through | 113 |
| abstract_inverted_index.trained | 61, 155 |
| abstract_inverted_index.without | 144 |
| abstract_inverted_index.Language | 76 |
| abstract_inverted_index.approach | 118 |
| abstract_inverted_index.attacks, | 125 |
| abstract_inverted_index.attacks. | 92 |
| abstract_inverted_index.caption, | 102 |
| abstract_inverted_index.existing | 140 |
| abstract_inverted_index.improved | 31 |
| abstract_inverted_index.indicate | 135 |
| abstract_inverted_index.network. | 116 |
| abstract_inverted_index.networks | 29 |
| abstract_inverted_index.overcome | 67 |
| abstract_inverted_index.requires | 51 |
| abstract_inverted_index.specific | 52 |
| abstract_inverted_index.training | 50 |
| abstract_inverted_index.vectors, | 56 |
| abstract_inverted_index.datasets, | 158 |
| abstract_inverted_index.diffusion | 26, 83, 115 |
| abstract_inverted_index.direction | 162 |
| abstract_inverted_index.efficient | 41 |
| abstract_inverted_index.enhancing | 130 |
| abstract_inverted_index.evaluated | 121 |
| abstract_inverted_index.examples. | 65 |
| abstract_inverted_index.generates | 100 |
| abstract_inverted_index.introduce | 71 |
| abstract_inverted_index.knowledge | 53 |
| abstract_inverted_index.promising | 161 |
| abstract_inverted_index.requiring | 145 |
| abstract_inverted_index.research. | 165 |
| abstract_inverted_index.training. | 148 |
| abstract_inverted_index.utilizing | 81 |
| abstract_inverted_index.versatile | 18 |
| abstract_inverted_index.classifier | 13 |
| abstract_inverted_index.extension, | 36 |
| abstract_inverted_index.framework, | 74 |
| abstract_inverted_index.generation | 33 |
| abstract_inverted_index.generative | 3 |
| abstract_inverted_index.generators | 87 |
| abstract_inverted_index.intensive. | 22 |
| abstract_inverted_index.techniques | 143 |
| abstract_inverted_index.Adversarial | 0, 78 |
| abstract_inverted_index.adversarial | 7, 37, 44, 49, 64, 91, 110, 124, 131, 141 |
| abstract_inverted_index.extensively | 62 |
| abstract_inverted_index.outperforms | 138 |
| abstract_inverted_index.pre-trained | 82 |
| abstract_inverted_index.robustness. | 132 |
| abstract_inverted_index.specialized | 146 |
| abstract_inverted_index.underscores | 150 |
| abstract_inverted_index.Purification | 79 |
| abstract_inverted_index.demonstrates | 5 |
| abstract_inverted_index.highlighting | 159 |
| abstract_inverted_index.limitations, | 69 |
| abstract_inverted_index.performance. | 9 |
| abstract_inverted_index.purification | 1, 111 |
| abstract_inverted_index.effectiveness | 128 |
| abstract_inverted_index.purification. | 38 |
| abstract_inverted_index.computationally | 21 |
| abstract_inverted_index.attack-agnostic, | 15 |
| abstract_inverted_index.generalizability | 152 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 2 |
| sustainable_development_goals[0].id | https://metadata.un.org/sdg/16 |
| sustainable_development_goals[0].score | 0.46000000834465027 |
| sustainable_development_goals[0].display_name | Peace, Justice and strong institutions |
| citation_normalized_percentile |