Err on the Side of Texture: Texture Bias on Real Data Article Swipe
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2412.10597
Bias significantly undermines both the accuracy and trustworthiness of machine learning models. To date, one of the strongest biases observed in image classification models is texture bias-where models overly rely on texture information rather than shape information. Yet, existing approaches for measuring and mitigating texture bias have not been able to capture how textures impact model robustness in real-world settings. In this work, we introduce the Texture Association Value (TAV), a novel metric that quantifies how strongly models rely on the presence of specific textures when classifying objects. Leveraging TAV, we demonstrate that model accuracy and robustness are heavily influenced by texture. Our results show that texture bias explains the existence of natural adversarial examples, where over 90% of these samples contain textures that are misaligned with the learned texture of their true label, resulting in confident mispredictions.
Related Topics
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2412.10597
- https://arxiv.org/pdf/2412.10597
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4405468204
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4405468204Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2412.10597Digital Object Identifier
- Title
-
Err on the Side of Texture: Texture Bias on Real DataWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2024Year of publication
- Publication date
-
2024-12-13Full publication date if available
- Authors
-
Blaine Hoak, Ryan Sheatsley, Patrick McDanielList of authors in order
- Landing page
-
https://arxiv.org/abs/2412.10597Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2412.10597Direct 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/2412.10597Direct OA link when available
- Concepts
-
Texture (cosmology), Computer science, Artificial intelligence, Pattern recognition (psychology), Image (mathematics)Top 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/W4405468204 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2412.10597 |
| ids.doi | https://doi.org/10.48550/arxiv.2412.10597 |
| ids.openalex | https://openalex.org/W4405468204 |
| fwci | |
| type | preprint |
| title | Err on the Side of Texture: Texture Bias on Real Data |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T12422 |
| topics[0].field.id | https://openalex.org/fields/27 |
| topics[0].field.display_name | Medicine |
| topics[0].score | 0.392300009727478 |
| topics[0].domain.id | https://openalex.org/domains/4 |
| topics[0].domain.display_name | Health Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/2741 |
| topics[0].subfield.display_name | Radiology, Nuclear Medicine and Imaging |
| topics[0].display_name | Radiomics and Machine Learning in Medical Imaging |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C2781195486 |
| concepts[0].level | 3 |
| concepts[0].score | 0.7667877674102783 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q289436 |
| concepts[0].display_name | Texture (cosmology) |
| concepts[1].id | https://openalex.org/C41008148 |
| concepts[1].level | 0 |
| concepts[1].score | 0.36392080783843994 |
| 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 | 0.36018460988998413 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[2].display_name | Artificial intelligence |
| concepts[3].id | https://openalex.org/C153180895 |
| concepts[3].level | 2 |
| concepts[3].score | 0.3490743041038513 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q7148389 |
| concepts[3].display_name | Pattern recognition (psychology) |
| concepts[4].id | https://openalex.org/C115961682 |
| concepts[4].level | 2 |
| concepts[4].score | 0.09259900450706482 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q860623 |
| concepts[4].display_name | Image (mathematics) |
| keywords[0].id | https://openalex.org/keywords/texture |
| keywords[0].score | 0.7667877674102783 |
| keywords[0].display_name | Texture (cosmology) |
| keywords[1].id | https://openalex.org/keywords/computer-science |
| keywords[1].score | 0.36392080783843994 |
| keywords[1].display_name | Computer science |
| keywords[2].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[2].score | 0.36018460988998413 |
| keywords[2].display_name | Artificial intelligence |
| keywords[3].id | https://openalex.org/keywords/pattern-recognition |
| keywords[3].score | 0.3490743041038513 |
| keywords[3].display_name | Pattern recognition (psychology) |
| keywords[4].id | https://openalex.org/keywords/image |
| keywords[4].score | 0.09259900450706482 |
| keywords[4].display_name | Image (mathematics) |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2412.10597 |
| 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/2412.10597 |
| 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/2412.10597 |
| locations[1].id | doi:10.48550/arxiv.2412.10597 |
| 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.2412.10597 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5088915686 |
| authorships[0].author.orcid | https://orcid.org/0000-0003-2960-0686 |
| authorships[0].author.display_name | Blaine Hoak |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Hoak, Blaine |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5056794879 |
| authorships[1].author.orcid | https://orcid.org/0000-0001-8447-602X |
| authorships[1].author.display_name | Ryan Sheatsley |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Sheatsley, Ryan |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5055368149 |
| authorships[2].author.orcid | https://orcid.org/0000-0003-2091-7484 |
| authorships[2].author.display_name | Patrick McDaniel |
| authorships[2].author_position | last |
| authorships[2].raw_author_name | McDaniel, Patrick |
| authorships[2].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/2412.10597 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Err on the Side of Texture: Texture Bias on Real Data |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T12422 |
| primary_topic.field.id | https://openalex.org/fields/27 |
| primary_topic.field.display_name | Medicine |
| primary_topic.score | 0.392300009727478 |
| primary_topic.domain.id | https://openalex.org/domains/4 |
| primary_topic.domain.display_name | Health Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/2741 |
| primary_topic.subfield.display_name | Radiology, Nuclear Medicine and Imaging |
| primary_topic.display_name | Radiomics and Machine Learning in Medical Imaging |
| related_works | https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2390279801, https://openalex.org/W4391913857, https://openalex.org/W2358668433, https://openalex.org/W4396701345, https://openalex.org/W2376932109, https://openalex.org/W2033914206, https://openalex.org/W2042327336 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2412.10597 |
| 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/2412.10597 |
| 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/2412.10597 |
| primary_location.id | pmh:oai:arXiv.org:2412.10597 |
| 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/2412.10597 |
| 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/2412.10597 |
| publication_date | 2024-12-13 |
| publication_year | 2024 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 70 |
| abstract_inverted_index.In | 60 |
| abstract_inverted_index.To | 12 |
| abstract_inverted_index.by | 100 |
| abstract_inverted_index.in | 20, 57, 135 |
| abstract_inverted_index.is | 24 |
| abstract_inverted_index.of | 8, 15, 82, 111, 118, 130 |
| abstract_inverted_index.on | 30, 79 |
| abstract_inverted_index.to | 50 |
| abstract_inverted_index.we | 63, 90 |
| abstract_inverted_index.90% | 117 |
| abstract_inverted_index.Our | 102 |
| abstract_inverted_index.and | 6, 42, 95 |
| abstract_inverted_index.are | 97, 124 |
| abstract_inverted_index.for | 40 |
| abstract_inverted_index.how | 52, 75 |
| abstract_inverted_index.not | 47 |
| abstract_inverted_index.one | 14 |
| abstract_inverted_index.the | 4, 16, 65, 80, 109, 127 |
| abstract_inverted_index.Bias | 0 |
| abstract_inverted_index.TAV, | 89 |
| abstract_inverted_index.Yet, | 37 |
| abstract_inverted_index.able | 49 |
| abstract_inverted_index.been | 48 |
| abstract_inverted_index.bias | 45, 107 |
| abstract_inverted_index.both | 3 |
| abstract_inverted_index.have | 46 |
| abstract_inverted_index.over | 116 |
| abstract_inverted_index.rely | 29, 78 |
| abstract_inverted_index.show | 104 |
| abstract_inverted_index.than | 34 |
| abstract_inverted_index.that | 73, 92, 105, 123 |
| abstract_inverted_index.this | 61 |
| abstract_inverted_index.true | 132 |
| abstract_inverted_index.when | 85 |
| abstract_inverted_index.with | 126 |
| abstract_inverted_index.Value | 68 |
| abstract_inverted_index.date, | 13 |
| abstract_inverted_index.image | 21 |
| abstract_inverted_index.model | 55, 93 |
| abstract_inverted_index.novel | 71 |
| abstract_inverted_index.shape | 35 |
| abstract_inverted_index.their | 131 |
| abstract_inverted_index.these | 119 |
| abstract_inverted_index.where | 115 |
| abstract_inverted_index.work, | 62 |
| abstract_inverted_index.(TAV), | 69 |
| abstract_inverted_index.biases | 18 |
| abstract_inverted_index.impact | 54 |
| abstract_inverted_index.label, | 133 |
| abstract_inverted_index.metric | 72 |
| abstract_inverted_index.models | 23, 27, 77 |
| abstract_inverted_index.overly | 28 |
| abstract_inverted_index.rather | 33 |
| abstract_inverted_index.Texture | 66 |
| abstract_inverted_index.capture | 51 |
| abstract_inverted_index.contain | 121 |
| abstract_inverted_index.heavily | 98 |
| abstract_inverted_index.learned | 128 |
| abstract_inverted_index.machine | 9 |
| abstract_inverted_index.models. | 11 |
| abstract_inverted_index.natural | 112 |
| abstract_inverted_index.results | 103 |
| abstract_inverted_index.samples | 120 |
| abstract_inverted_index.texture | 25, 31, 44, 106, 129 |
| abstract_inverted_index.accuracy | 5, 94 |
| abstract_inverted_index.existing | 38 |
| abstract_inverted_index.explains | 108 |
| abstract_inverted_index.learning | 10 |
| abstract_inverted_index.objects. | 87 |
| abstract_inverted_index.observed | 19 |
| abstract_inverted_index.presence | 81 |
| abstract_inverted_index.specific | 83 |
| abstract_inverted_index.strongly | 76 |
| abstract_inverted_index.texture. | 101 |
| abstract_inverted_index.textures | 53, 84, 122 |
| abstract_inverted_index.confident | 136 |
| abstract_inverted_index.examples, | 114 |
| abstract_inverted_index.existence | 110 |
| abstract_inverted_index.introduce | 64 |
| abstract_inverted_index.measuring | 41 |
| abstract_inverted_index.resulting | 134 |
| abstract_inverted_index.settings. | 59 |
| abstract_inverted_index.strongest | 17 |
| abstract_inverted_index.Leveraging | 88 |
| abstract_inverted_index.approaches | 39 |
| abstract_inverted_index.bias-where | 26 |
| abstract_inverted_index.influenced | 99 |
| abstract_inverted_index.misaligned | 125 |
| abstract_inverted_index.mitigating | 43 |
| abstract_inverted_index.quantifies | 74 |
| abstract_inverted_index.real-world | 58 |
| abstract_inverted_index.robustness | 56, 96 |
| abstract_inverted_index.undermines | 2 |
| abstract_inverted_index.Association | 67 |
| abstract_inverted_index.adversarial | 113 |
| abstract_inverted_index.classifying | 86 |
| abstract_inverted_index.demonstrate | 91 |
| abstract_inverted_index.information | 32 |
| abstract_inverted_index.information. | 36 |
| abstract_inverted_index.significantly | 1 |
| abstract_inverted_index.classification | 22 |
| abstract_inverted_index.mispredictions. | 137 |
| abstract_inverted_index.trustworthiness | 7 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 3 |
| citation_normalized_percentile |