LeX-Art: Rethinking Text Generation via Scalable High-Quality Data Synthesis Article Swipe
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2503.21749
We introduce LeX-Art, a comprehensive suite for high-quality text-image synthesis that systematically bridges the gap between prompt expressiveness and text rendering fidelity. Our approach follows a data-centric paradigm, constructing a high-quality data synthesis pipeline based on Deepseek-R1 to curate LeX-10K, a dataset of 10K high-resolution, aesthetically refined 1024$\times$1024 images. Beyond dataset construction, we develop LeX-Enhancer, a robust prompt enrichment model, and train two text-to-image models, LeX-FLUX and LeX-Lumina, achieving state-of-the-art text rendering performance. To systematically evaluate visual text generation, we introduce LeX-Bench, a benchmark that assesses fidelity, aesthetics, and alignment, complemented by Pairwise Normalized Edit Distance (PNED), a novel metric for robust text accuracy evaluation. Experiments demonstrate significant improvements, with LeX-Lumina achieving a 79.81% PNED gain on CreateBench, and LeX-FLUX outperforming baselines in color (+3.18%), positional (+4.45%), and font accuracy (+3.81%). Our codes, models, datasets, and demo are publicly available.
Related Topics
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2503.21749
- https://arxiv.org/pdf/2503.21749
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W4415063940
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4415063940Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2503.21749Digital Object Identifier
- Title
-
LeX-Art: Rethinking Text Generation via Scalable High-Quality Data SynthesisWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-03-27Full publication date if available
- Authors
-
Shitian Zhao, Qilong Wu, Xinyue Li, Bo Zhang, Ming D. Li, Qi Qin, Dongyang Liu, Kaipeng Zhang, Hongsheng Li, Yu Qiao, Peng Gao, Bin Fu, Zhen LiList of authors in order
- Landing page
-
https://arxiv.org/abs/2503.21749Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2503.21749Direct 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/2503.21749Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4415063940 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2503.21749 |
| ids.doi | https://doi.org/10.48550/arxiv.2503.21749 |
| ids.openalex | https://openalex.org/W4415063940 |
| fwci | |
| type | preprint |
| title | LeX-Art: Rethinking Text Generation via Scalable High-Quality Data Synthesis |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10181 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.8428999781608582 |
| 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 | Natural Language Processing Techniques |
| topics[1].id | https://openalex.org/T14339 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.7674000263214111 |
| 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 | Image Processing and 3D Reconstruction |
| topics[2].id | https://openalex.org/T13523 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.7544999718666077 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1703 |
| topics[2].subfield.display_name | Computational Theory and Mathematics |
| topics[2].display_name | Mathematics, Computing, and Information Processing |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2503.21749 |
| 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/2503.21749 |
| 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/2503.21749 |
| locations[1].id | doi:10.48550/arxiv.2503.21749 |
| 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.2503.21749 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5102591920 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Shitian Zhao |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Zhao, Shitian |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5004367003 |
| authorships[1].author.orcid | |
| authorships[1].author.display_name | Qilong Wu |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Wu, Qilong |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5100357340 |
| authorships[2].author.orcid | https://orcid.org/0009-0000-9238-0458 |
| authorships[2].author.display_name | Xinyue Li |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Li, Xinyue |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5077920649 |
| authorships[3].author.orcid | https://orcid.org/0000-0001-7797-0038 |
| authorships[3].author.display_name | Bo Zhang |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Zhang, Bo |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5005268084 |
| authorships[4].author.orcid | https://orcid.org/0000-0002-6873-9472 |
| authorships[4].author.display_name | Ming D. Li |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Li, Ming |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5092827823 |
| authorships[5].author.orcid | https://orcid.org/0009-0003-2445-7814 |
| authorships[5].author.display_name | Qi Qin |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Qin, Qi |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5100761788 |
| authorships[6].author.orcid | https://orcid.org/0000-0002-0608-8192 |
| authorships[6].author.display_name | Dongyang Liu |
| authorships[6].author_position | middle |
| authorships[6].raw_author_name | Liu, Dongyang |
| authorships[6].is_corresponding | False |
| authorships[7].author.id | https://openalex.org/A5036606244 |
| authorships[7].author.orcid | https://orcid.org/0000-0001-6105-6532 |
| authorships[7].author.display_name | Kaipeng Zhang |
| authorships[7].author_position | middle |
| authorships[7].raw_author_name | Zhang, Kaipeng |
| authorships[7].is_corresponding | False |
| authorships[8].author.id | https://openalex.org/A5114506692 |
| authorships[8].author.orcid | https://orcid.org/0009-0003-4821-5051 |
| authorships[8].author.display_name | Hongsheng Li |
| authorships[8].author_position | middle |
| authorships[8].raw_author_name | Li, Hongsheng |
| authorships[8].is_corresponding | False |
| authorships[9].author.id | https://openalex.org/A5100748135 |
| authorships[9].author.orcid | https://orcid.org/0000-0002-1889-2567 |
| authorships[9].author.display_name | Yu Qiao |
| authorships[9].author_position | middle |
| authorships[9].raw_author_name | Qiao, Yu |
| authorships[9].is_corresponding | False |
| authorships[10].author.id | https://openalex.org/A5028272399 |
| authorships[10].author.orcid | https://orcid.org/0000-0003-2230-3937 |
| authorships[10].author.display_name | Peng Gao |
| authorships[10].author_position | middle |
| authorships[10].raw_author_name | Gao, Peng |
| authorships[10].is_corresponding | False |
| authorships[11].author.id | https://openalex.org/A5056461571 |
| authorships[11].author.orcid | https://orcid.org/0000-0002-8352-9393 |
| authorships[11].author.display_name | Bin Fu |
| authorships[11].author_position | middle |
| authorships[11].raw_author_name | Fu, Bin |
| authorships[11].is_corresponding | False |
| authorships[12].author.id | https://openalex.org/A5100332625 |
| authorships[12].author.orcid | https://orcid.org/0000-0002-7669-2686 |
| authorships[12].author.display_name | Zhen Li |
| authorships[12].author_position | last |
| authorships[12].raw_author_name | Li, Zhen |
| authorships[12].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/2503.21749 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-11T00:00:00 |
| display_name | LeX-Art: Rethinking Text Generation via Scalable High-Quality Data Synthesis |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T10181 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.8428999781608582 |
| 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 | Natural Language Processing Techniques |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2503.21749 |
| 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/2503.21749 |
| 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/2503.21749 |
| primary_location.id | pmh:oai:arXiv.org:2503.21749 |
| 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/2503.21749 |
| 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/2503.21749 |
| publication_date | 2025-03-27 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 3, 25, 29, 40, 55, 82, 97, 112 |
| abstract_inverted_index.To | 73 |
| abstract_inverted_index.We | 0 |
| abstract_inverted_index.by | 91 |
| abstract_inverted_index.in | 122 |
| abstract_inverted_index.of | 42 |
| abstract_inverted_index.on | 35, 116 |
| abstract_inverted_index.to | 37 |
| abstract_inverted_index.we | 52, 79 |
| abstract_inverted_index.10K | 43 |
| abstract_inverted_index.Our | 22, 131 |
| abstract_inverted_index.and | 18, 60, 66, 88, 118, 127, 135 |
| abstract_inverted_index.are | 137 |
| abstract_inverted_index.for | 6, 100 |
| abstract_inverted_index.gap | 14 |
| abstract_inverted_index.the | 13 |
| abstract_inverted_index.two | 62 |
| abstract_inverted_index.Edit | 94 |
| abstract_inverted_index.PNED | 114 |
| abstract_inverted_index.data | 31 |
| abstract_inverted_index.demo | 136 |
| abstract_inverted_index.font | 128 |
| abstract_inverted_index.gain | 115 |
| abstract_inverted_index.text | 19, 70, 77, 102 |
| abstract_inverted_index.that | 10, 84 |
| abstract_inverted_index.with | 109 |
| abstract_inverted_index.based | 34 |
| abstract_inverted_index.color | 123 |
| abstract_inverted_index.novel | 98 |
| abstract_inverted_index.suite | 5 |
| abstract_inverted_index.train | 61 |
| abstract_inverted_index.79.81% | 113 |
| abstract_inverted_index.Beyond | 49 |
| abstract_inverted_index.codes, | 132 |
| abstract_inverted_index.curate | 38 |
| abstract_inverted_index.metric | 99 |
| abstract_inverted_index.model, | 59 |
| abstract_inverted_index.prompt | 16, 57 |
| abstract_inverted_index.robust | 56, 101 |
| abstract_inverted_index.visual | 76 |
| abstract_inverted_index.(PNED), | 96 |
| abstract_inverted_index.between | 15 |
| abstract_inverted_index.bridges | 12 |
| abstract_inverted_index.dataset | 41, 50 |
| abstract_inverted_index.develop | 53 |
| abstract_inverted_index.follows | 24 |
| abstract_inverted_index.images. | 48 |
| abstract_inverted_index.models, | 64, 133 |
| abstract_inverted_index.refined | 46 |
| abstract_inverted_index.Distance | 95 |
| abstract_inverted_index.LeX-10K, | 39 |
| abstract_inverted_index.LeX-Art, | 2 |
| abstract_inverted_index.LeX-FLUX | 65, 119 |
| abstract_inverted_index.Pairwise | 92 |
| abstract_inverted_index.accuracy | 103, 129 |
| abstract_inverted_index.approach | 23 |
| abstract_inverted_index.assesses | 85 |
| abstract_inverted_index.evaluate | 75 |
| abstract_inverted_index.pipeline | 33 |
| abstract_inverted_index.publicly | 138 |
| abstract_inverted_index.(+3.18%), | 124 |
| abstract_inverted_index.(+3.81%). | 130 |
| abstract_inverted_index.(+4.45%), | 126 |
| abstract_inverted_index.achieving | 68, 111 |
| abstract_inverted_index.baselines | 121 |
| abstract_inverted_index.benchmark | 83 |
| abstract_inverted_index.datasets, | 134 |
| abstract_inverted_index.fidelity, | 86 |
| abstract_inverted_index.fidelity. | 21 |
| abstract_inverted_index.introduce | 1, 80 |
| abstract_inverted_index.paradigm, | 27 |
| abstract_inverted_index.rendering | 20, 71 |
| abstract_inverted_index.synthesis | 9, 32 |
| abstract_inverted_index.LeX-Bench, | 81 |
| abstract_inverted_index.LeX-Lumina | 110 |
| abstract_inverted_index.Normalized | 93 |
| abstract_inverted_index.alignment, | 89 |
| abstract_inverted_index.available. | 139 |
| abstract_inverted_index.enrichment | 58 |
| abstract_inverted_index.positional | 125 |
| abstract_inverted_index.text-image | 8 |
| abstract_inverted_index.Deepseek-R1 | 36 |
| abstract_inverted_index.Experiments | 105 |
| abstract_inverted_index.LeX-Lumina, | 67 |
| abstract_inverted_index.aesthetics, | 87 |
| abstract_inverted_index.demonstrate | 106 |
| abstract_inverted_index.evaluation. | 104 |
| abstract_inverted_index.generation, | 78 |
| abstract_inverted_index.significant | 107 |
| abstract_inverted_index.CreateBench, | 117 |
| abstract_inverted_index.complemented | 90 |
| abstract_inverted_index.constructing | 28 |
| abstract_inverted_index.data-centric | 26 |
| abstract_inverted_index.high-quality | 7, 30 |
| abstract_inverted_index.performance. | 72 |
| abstract_inverted_index.LeX-Enhancer, | 54 |
| abstract_inverted_index.aesthetically | 45 |
| abstract_inverted_index.comprehensive | 4 |
| abstract_inverted_index.construction, | 51 |
| abstract_inverted_index.improvements, | 108 |
| abstract_inverted_index.outperforming | 120 |
| abstract_inverted_index.text-to-image | 63 |
| abstract_inverted_index.expressiveness | 17 |
| abstract_inverted_index.systematically | 11, 74 |
| abstract_inverted_index.1024$\times$1024 | 47 |
| abstract_inverted_index.high-resolution, | 44 |
| abstract_inverted_index.state-of-the-art | 69 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 13 |
| citation_normalized_percentile |