{"id":"https://openalex.org/W7138346166","doi":"https://doi.org/10.48550/arxiv.2603.13389","title":"High-Fidelity Text-to-Image Generation from Pre-Trained Vision-Language Models via Distribution-Conditioned Diffusion Decoding","display_name":"High-Fidelity Text-to-Image Generation from Pre-Trained Vision-Language Models via Distribution-Conditioned Diffusion Decoding","publication_year":2026,"publication_date":"2026-03-11","ids":{"openalex":"https://openalex.org/W7138346166","doi":"https://doi.org/10.48550/arxiv.2603.13389"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.13389","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.13389","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.13389","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5046143080","display_name":"Ji Woo Hong","orcid":"https://orcid.org/0000-0002-3758-0307"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hong, Ji Woo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103277375","display_name":"Hee Suk Yoon","orcid":"https://orcid.org/0000-0003-2115-8459"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yoon, Hee Suk","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093047686","display_name":"Gwanhyeong Koo","orcid":"https://orcid.org/0009-0005-6455-3223"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Koo, Gwanhyeong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031585399","display_name":"Eunseop Yoon","orcid":"https://orcid.org/0000-0002-5580-5354"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yoon, Eunseop","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108835885","display_name":"Soohwan Eom","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Eom, SooHwan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129705563","display_name":"Qi Dai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dai, Qi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129641229","display_name":"Chong Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Chong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129682344","display_name":"Chang D. Yoo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yoo, Chang D.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5773000121116638,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5773000121116638,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.249099999666214,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.01889999955892563,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.720300018787384},{"id":"https://openalex.org/keywords/fidelity","display_name":"Fidelity","score":0.6331999897956848},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5773000121116638},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4661000072956085},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.4399999976158142},{"id":"https://openalex.org/keywords/high-fidelity","display_name":"High fidelity","score":0.3840000033378601},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.36010000109672546},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.32199999690055847}],"concepts":[{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.720300018787384},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.652400016784668},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.6331999897956848},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5773000121116638},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4778999984264374},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4661000072956085},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45719999074935913},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4526999890804291},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.4399999976158142},{"id":"https://openalex.org/C113364801","wikidata":"https://www.wikidata.org/wiki/Q26674","display_name":"High fidelity","level":2,"score":0.3840000033378601},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.36010000109672546},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.32199999690055847},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.32100000977516174},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3188000023365021},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.2678000032901764},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.26100000739097595},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.2587999999523163},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C141379421","wikidata":"https://www.wikidata.org/wiki/Q6094427","display_name":"Iterative reconstruction","level":2,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.13389","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.13389","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.13389","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.13389","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"large-scale":[1,47],"vision-language":[2],"models":[3,42],"(VLMs)":[4],"have":[5,30],"shown":[6],"remarkable":[7],"text-to-image":[8,166],"generation":[9],"capabilities,":[10],"yet":[11],"their":[12],"visual":[13,37,159],"fidelity":[14,70,160],"remains":[15],"constrained":[16],"by":[17,71],"the":[18,54,78,87,98,127,134,141],"discrete":[19],"image":[20,69],"tokenization,":[21],"which":[22],"poses":[23],"a":[24,63,74],"major":[25],"challenge.":[26],"Although":[27],"several":[28],"studies":[29],"explored":[31],"continuous":[32],"representation":[33],"modeling":[34],"to":[35,43,53],"enhance":[36],"quality,":[38],"adapting":[39],"pre-trained":[40,83],"VLM":[41],"such":[44],"representations":[45],"requires":[46],"data":[48],"and":[49,165],"training":[50,72,152],"costs":[51],"comparable":[52],"original":[55,88],"pre-training.":[56],"To":[57],"circumvent":[58],"this":[59],"limitation,":[60],"we":[61],"propose":[62],"diffusion-based":[64],"decoding":[65],"framework":[66],"that":[67],"enhances":[68],"only":[73],"diffusion":[75,116],"decoder":[76],"on":[77,138,153],"output":[79],"image-token":[80,100],"logits":[81,101,125],"of":[82],"VLMs,":[84],"thereby":[85],"preserving":[86],"model":[89],"intact.":[90],"At":[91],"its":[92],"core,":[93],"Logit-to-Code":[94],"Distributional":[95],"Mapping":[96],"converts":[97],"VLM's":[99],"into":[102],"continuous,":[103],"distribution-weighted":[104],"code":[105],"vectors":[106],"with":[107,130],"uncertainty":[108],"features,":[109],"providing":[110],"an":[111],"effective":[112],"conditioning":[113],"signal":[114],"for":[115,161],"decoding.":[117],"A":[118],"lightweight":[119],"Logit":[120],"Calibration":[121],"aligns":[122],"training-time":[123],"proxy":[124],"from":[126,168],"VQ-VAE":[128,163],"encoder":[129],"VLM-generated":[131],"logits,":[132],"mitigating":[133],"train-inference":[135],"gap.":[136],"Conditioned":[137],"these":[139],"representations,":[140],"Distribution-Conditioned":[142],"Diffusion":[143],"Decoder":[144],"generates":[145],"high-fidelity":[146],"images.":[147],"Achieved":[148],"solely":[149],"through":[150],"short":[151],"ImageNet-1K,":[154],"our":[155],"method":[156],"consistently":[157],"improves":[158],"both":[162],"reconstructions":[164],"generations":[167],"VLM-predicted":[169],"tokens.":[170]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-18T00:00:00"}