Concept-Centric Token Interpretation for Vector-Quantized Generative Models Article Swipe
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2506.00698
Vector-Quantized Generative Models (VQGMs) have emerged as powerful tools for image generation. However, the key component of VQGMs -- the codebook of discrete tokens -- is still not well understood, e.g., which tokens are critical to generate an image of a certain concept? This paper introduces Concept-Oriented Token Explanation (CORTEX), a novel approach for interpreting VQGMs by identifying concept-specific token combinations. Our framework employs two methods: (1) a sample-level explanation method that analyzes token importance scores in individual images, and (2) a codebook-level explanation method that explores the entire codebook to find globally relevant tokens. Experimental results demonstrate CORTEX's efficacy in providing clear explanations of token usage in the generative process, outperforming baselines across multiple pretrained VQGMs. Besides enhancing VQGMs transparency, CORTEX is useful in applications such as targeted image editing and shortcut feature detection. Our code is available at https://github.com/YangTianze009/CORTEX.
Related Topics
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2506.00698
- https://arxiv.org/pdf/2506.00698
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W4414891767
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4414891767Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2506.00698Digital Object Identifier
- Title
-
Concept-Centric Token Interpretation for Vector-Quantized Generative ModelsWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-05-31Full publication date if available
- Authors
-
Tianze Yang, Yucheng Shi, Mengnan Du, Xuansheng Wu, Qiaoyu Tan, Jing Sun, Ninghao LiuList of authors in order
- Landing page
-
https://arxiv.org/abs/2506.00698Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2506.00698Direct 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/2506.00698Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4414891767 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2506.00698 |
| ids.doi | https://doi.org/10.48550/arxiv.2506.00698 |
| ids.openalex | https://openalex.org/W4414891767 |
| fwci | |
| type | preprint |
| title | Concept-Centric Token Interpretation for Vector-Quantized Generative Models |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10215 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9860000014305115 |
| 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 | Semantic Web and Ontologies |
| topics[1].id | https://openalex.org/T10317 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9416000247001648 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1705 |
| topics[1].subfield.display_name | Computer Networks and Communications |
| topics[1].display_name | Advanced Database Systems and Queries |
| topics[2].id | https://openalex.org/T11986 |
| topics[2].field.id | https://openalex.org/fields/18 |
| topics[2].field.display_name | Decision Sciences |
| topics[2].score | 0.9332000017166138 |
| topics[2].domain.id | https://openalex.org/domains/2 |
| topics[2].domain.display_name | Social Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1802 |
| topics[2].subfield.display_name | Information Systems and Management |
| topics[2].display_name | Scientific Computing and Data Management |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2506.00698 |
| 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/2506.00698 |
| 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/2506.00698 |
| locations[1].id | doi:10.48550/arxiv.2506.00698 |
| 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.2506.00698 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5101301594 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Tianze Yang |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Yang, Tianze |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5016550690 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-8070-5363 |
| authorships[1].author.display_name | Yucheng Shi |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Shi, Yucheng |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5072191151 |
| authorships[2].author.orcid | https://orcid.org/0000-0002-1614-6069 |
| authorships[2].author.display_name | Mengnan Du |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Du, Mengnan |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5089884284 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-7816-7658 |
| authorships[3].author.display_name | Xuansheng Wu |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Wu, Xuansheng |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5102981159 |
| authorships[4].author.orcid | https://orcid.org/0000-0002-1206-8446 |
| authorships[4].author.display_name | Qiaoyu Tan |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Tan, Qiaoyu |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5084282975 |
| authorships[5].author.orcid | https://orcid.org/0000-0001-7254-454X |
| authorships[5].author.display_name | Jing Sun |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Sun, Jin |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5007489034 |
| authorships[6].author.orcid | https://orcid.org/0000-0002-9170-2424 |
| authorships[6].author.display_name | Ninghao Liu |
| authorships[6].author_position | last |
| authorships[6].raw_author_name | Liu, Ninghao |
| authorships[6].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/2506.00698 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Concept-Centric Token Interpretation for Vector-Quantized Generative Models |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T10215 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9860000014305115 |
| 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 | Semantic Web and Ontologies |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2506.00698 |
| 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/2506.00698 |
| 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/2506.00698 |
| primary_location.id | pmh:oai:arXiv.org:2506.00698 |
| 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/2506.00698 |
| 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/2506.00698 |
| publication_date | 2025-05-31 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 40, 50, 67, 81 |
| abstract_inverted_index.-- | 18, 24 |
| abstract_inverted_index.an | 37 |
| abstract_inverted_index.as | 6, 127 |
| abstract_inverted_index.at | 139 |
| abstract_inverted_index.by | 56 |
| abstract_inverted_index.in | 76, 100, 107, 124 |
| abstract_inverted_index.is | 25, 122, 137 |
| abstract_inverted_index.of | 16, 21, 39, 104 |
| abstract_inverted_index.to | 35, 90 |
| abstract_inverted_index.(1) | 66 |
| abstract_inverted_index.(2) | 80 |
| abstract_inverted_index.Our | 61, 135 |
| abstract_inverted_index.and | 79, 131 |
| abstract_inverted_index.are | 33 |
| abstract_inverted_index.for | 9, 53 |
| abstract_inverted_index.key | 14 |
| abstract_inverted_index.not | 27 |
| abstract_inverted_index.the | 13, 19, 87, 108 |
| abstract_inverted_index.two | 64 |
| abstract_inverted_index.This | 43 |
| abstract_inverted_index.code | 136 |
| abstract_inverted_index.find | 91 |
| abstract_inverted_index.have | 4 |
| abstract_inverted_index.such | 126 |
| abstract_inverted_index.that | 71, 85 |
| abstract_inverted_index.well | 28 |
| abstract_inverted_index.Token | 47 |
| abstract_inverted_index.VQGMs | 17, 55, 119 |
| abstract_inverted_index.clear | 102 |
| abstract_inverted_index.e.g., | 30 |
| abstract_inverted_index.image | 10, 38, 129 |
| abstract_inverted_index.novel | 51 |
| abstract_inverted_index.paper | 44 |
| abstract_inverted_index.still | 26 |
| abstract_inverted_index.token | 59, 73, 105 |
| abstract_inverted_index.tools | 8 |
| abstract_inverted_index.usage | 106 |
| abstract_inverted_index.which | 31 |
| abstract_inverted_index.CORTEX | 121 |
| abstract_inverted_index.Models | 2 |
| abstract_inverted_index.VQGMs. | 116 |
| abstract_inverted_index.across | 113 |
| abstract_inverted_index.entire | 88 |
| abstract_inverted_index.method | 70, 84 |
| abstract_inverted_index.scores | 75 |
| abstract_inverted_index.tokens | 23, 32 |
| abstract_inverted_index.useful | 123 |
| abstract_inverted_index.(VQGMs) | 3 |
| abstract_inverted_index.Besides | 117 |
| abstract_inverted_index.certain | 41 |
| abstract_inverted_index.editing | 130 |
| abstract_inverted_index.emerged | 5 |
| abstract_inverted_index.employs | 63 |
| abstract_inverted_index.feature | 133 |
| abstract_inverted_index.images, | 78 |
| abstract_inverted_index.results | 96 |
| abstract_inverted_index.tokens. | 94 |
| abstract_inverted_index.CORTEX's | 98 |
| abstract_inverted_index.However, | 12 |
| abstract_inverted_index.analyzes | 72 |
| abstract_inverted_index.approach | 52 |
| abstract_inverted_index.codebook | 20, 89 |
| abstract_inverted_index.concept? | 42 |
| abstract_inverted_index.critical | 34 |
| abstract_inverted_index.discrete | 22 |
| abstract_inverted_index.efficacy | 99 |
| abstract_inverted_index.explores | 86 |
| abstract_inverted_index.generate | 36 |
| abstract_inverted_index.globally | 92 |
| abstract_inverted_index.methods: | 65 |
| abstract_inverted_index.multiple | 114 |
| abstract_inverted_index.powerful | 7 |
| abstract_inverted_index.process, | 110 |
| abstract_inverted_index.relevant | 93 |
| abstract_inverted_index.shortcut | 132 |
| abstract_inverted_index.targeted | 128 |
| abstract_inverted_index.(CORTEX), | 49 |
| abstract_inverted_index.available | 138 |
| abstract_inverted_index.baselines | 112 |
| abstract_inverted_index.component | 15 |
| abstract_inverted_index.enhancing | 118 |
| abstract_inverted_index.framework | 62 |
| abstract_inverted_index.providing | 101 |
| abstract_inverted_index.Generative | 1 |
| abstract_inverted_index.detection. | 134 |
| abstract_inverted_index.generative | 109 |
| abstract_inverted_index.importance | 74 |
| abstract_inverted_index.individual | 77 |
| abstract_inverted_index.introduces | 45 |
| abstract_inverted_index.pretrained | 115 |
| abstract_inverted_index.Explanation | 48 |
| abstract_inverted_index.demonstrate | 97 |
| abstract_inverted_index.explanation | 69, 83 |
| abstract_inverted_index.generation. | 11 |
| abstract_inverted_index.identifying | 57 |
| abstract_inverted_index.understood, | 29 |
| abstract_inverted_index.Experimental | 95 |
| abstract_inverted_index.applications | 125 |
| abstract_inverted_index.explanations | 103 |
| abstract_inverted_index.interpreting | 54 |
| abstract_inverted_index.sample-level | 68 |
| abstract_inverted_index.combinations. | 60 |
| abstract_inverted_index.outperforming | 111 |
| abstract_inverted_index.transparency, | 120 |
| abstract_inverted_index.codebook-level | 82 |
| abstract_inverted_index.Concept-Oriented | 46 |
| abstract_inverted_index.Vector-Quantized | 0 |
| abstract_inverted_index.concept-specific | 58 |
| abstract_inverted_index.https://github.com/YangTianze009/CORTEX. | 140 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 7 |
| citation_normalized_percentile |