{"id":"https://openalex.org/W4403580265","doi":"https://doi.org/10.48550/arxiv.2410.13863","title":"Fluid: Scaling Autoregressive Text-to-image Generative Models with Continuous Tokens","display_name":"Fluid: Scaling Autoregressive Text-to-image Generative Models with Continuous Tokens","publication_year":2024,"publication_date":"2024-10-17","ids":{"openalex":"https://openalex.org/W4403580265","doi":"https://doi.org/10.48550/arxiv.2410.13863"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2410.13863","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.13863","pdf_url":"https://arxiv.org/pdf/2410.13863","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2410.13863","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100537622","display_name":"Lijie Fan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fan, Lijie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101928475","display_name":"Tianhong Li","orcid":"https://orcid.org/0000-0001-6280-6411","observed_orcids":["https://orcid.org/0000-0001-6280-6411"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Tianhong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022572266","display_name":"Siyang Qin","orcid":"https://orcid.org/0000-0002-5776-1965","observed_orcids":["https://orcid.org/0000-0002-5776-1965"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qin, Siyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069912971","display_name":"Yuanzhen Li","orcid":"https://orcid.org/0000-0002-9831-8249","observed_orcids":["https://orcid.org/0000-0002-9831-8249","https://orcid.org/0000-0002-5514-9991"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yuanzhen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100722234","display_name":"Chen Sun","orcid":"https://orcid.org/0000-0001-8772-9627","observed_orcids":["https://orcid.org/0000-0001-8772-9627","https://orcid.org/0000-0002-3036-7212"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007681620","display_name":"Michael Rubinstein","orcid":"https://orcid.org/0000-0002-3707-3807","observed_orcids":["https://orcid.org/0000-0002-3707-3807"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rubinstein, Michael","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101440839","display_name":"Deqing Sun","orcid":"https://orcid.org/0000-0003-0329-0456","observed_orcids":["https://orcid.org/0000-0003-0329-0456"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Deqing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100700361","display_name":"Kaiming He","orcid":"https://orcid.org/0000-0001-7318-9658","observed_orcids":["https://orcid.org/0000-0001-7318-9658"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Kaiming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5085958820","display_name":"Yonglong Tian","orcid":"https://orcid.org/0000-0002-6110-2145","observed_orcids":["https://orcid.org/0000-0002-6110-2145"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tian, Yonglong","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":true,"cited_by_count":116,"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":0.9990000128746033,"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":0.9990000128746033,"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":0.0008999999845400453,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/text-to-image-generation","display_name":"text-to-image generation","score":0.9440000057220459},{"id":"https://openalex.org/keywords/autoregressive-image-generation","display_name":"autoregressive image generation","score":0.7670000195503235},{"id":"https://openalex.org/keywords/attention-mechanism","display_name":"attention mechanism","score":0.5059999823570251},{"id":"https://openalex.org/keywords/model-scaling","display_name":"model scaling","score":0.8050000071525574},{"id":"https://openalex.org/keywords/fluids","display_name":"fluids","score":0.5220000147819519},{"id":"https://openalex.org/keywords/fid","display_name":"FID","score":0.6850000023841858}],"concepts":[{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.8091508150100708},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6752200126647949},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.6153914928436279},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5876314043998718},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5302733182907104},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.44698381423950195},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3957333564758301},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3571093678474426},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.315993994474411},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2688731849193573},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.06901821494102478}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2410.13863","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.13863","pdf_url":"https://arxiv.org/pdf/2410.13863","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2410.13863","doi":"https://doi.org/10.48550/arxiv.2410.13863","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2410.13863","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2410.13863","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.13863","pdf_url":"https://arxiv.org/pdf/2410.13863","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403580265.pdf","grobid_xml":"https://content.openalex.org/works/W4403580265.grobid-xml"},"referenced_works_count":26,"referenced_works":["https://openalex.org/W2885185669","https://openalex.org/W2896457183","https://openalex.org/W2946948417","https://openalex.org/W2963799213","https://openalex.org/W3001279689","https://openalex.org/W3095645723","https://openalex.org/W3206395542","https://openalex.org/W4221167110","https://openalex.org/W4224035735","https://openalex.org/W4283026156","https://openalex.org/W4283388932","https://openalex.org/W4288099666","https://openalex.org/W4296406182","https://openalex.org/W4303440777","https://openalex.org/W4327810158","https://openalex.org/W4360836968","https://openalex.org/W4386556027","https://openalex.org/W4390041933","https://openalex.org/W4390963175","https://openalex.org/W4401306886","https://openalex.org/W4402427382","https://openalex.org/W4403012178","https://openalex.org/W4403662644","https://openalex.org/W4415796345","https://openalex.org/W4415797571","https://openalex.org/W7133225605"],"related_works":["https://openalex.org/W6910622637","https://openalex.org/W4283388932","https://openalex.org/W4399554455","https://openalex.org/W7129434047","https://openalex.org/W6979295168","https://openalex.org/W6929758863","https://openalex.org/W7160095423","https://openalex.org/W7128535520","https://openalex.org/W4416381181","https://openalex.org/W4414384363"],"abstract_inverted_index":{"Scaling":[0],"up":[1],"autoregressive":[2,138],"models":[3,36,67,119],"in":[4,12,24,47,70],"vision":[5,182],"has":[6],"not":[7],"proven":[8],"as":[9,11],"beneficial":[10],"large":[13],"language":[14,184],"models.":[15,128,185],"In":[16],"this":[17,21],"work,":[18],"we":[19,133],"investigate":[20],"scaling":[22,179],"problem":[23],"the":[25,107,115,162,178],"context":[26],"of":[27,72,152],"text-to-image":[28],"generation,":[29],"focusing":[30],"on":[31,93,140,154,161],"two":[32],"critical":[33],"factors:":[34],"whether":[35,43],"use":[37],"discrete":[38,104],"or":[39,50,56],"continuous":[40,94,141],"tokens,":[41],"and":[42,84,110,157,169,183],"tokens":[44,95],"are":[45],"generated":[46],"a":[48,136,147],"random":[49],"fixed":[51],"raster":[52],"order":[53,109],"using":[54,103],"BERT-":[55],"GPT-like":[57],"transformer":[58],"architectures.":[59],"Our":[60],"empirical":[61],"results":[62,170],"show":[63],"that,":[64],"while":[65],"all":[66],"scale":[68],"effectively":[69],"terms":[71],"validation":[73],"loss,":[74],"their":[75],"evaluation":[76],"performance":[77],"--":[78,87],"measured":[79],"by":[80,130],"FID,":[81],"GenEval":[82,116,123,163],"score,":[83],"visual":[85,99],"quality":[86,100],"follows":[88],"different":[89],"trends.":[90],"Models":[91],"based":[92],"achieve":[96,120],"significantly":[97,113],"better":[98,122],"than":[101],"those":[102],"tokens.":[105,142],"Furthermore,":[106],"generation":[108],"attention":[111],"mechanisms":[112],"affect":[114],"score:":[117],"random-order":[118,137],"notably":[121],"scores":[124],"compared":[125],"to":[126,175],"raster-order":[127],"Inspired":[129],"these":[131],"findings,":[132],"train":[134],"Fluid,":[135],"model":[139,145],"Fluid":[143],"10.5B":[144],"achieves":[146],"new":[148],"state-of-the-art":[149],"zero-shot":[150],"FID":[151],"6.16":[153],"MS-COCO":[155],"30K,":[156],"0.69":[158],"overall":[159],"score":[160],"benchmark.":[164],"We":[165],"hope":[166],"our":[167],"findings":[168],"will":[171],"encourage":[172],"future":[173],"efforts":[174],"further":[176],"bridge":[177],"gap":[180],"between":[181]},"counts_by_year":[{"year":2026,"cited_by_count":30},{"year":2025,"cited_by_count":77},{"year":2024,"cited_by_count":9}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2024-10-21T00:00:00"}