{"id":"https://openalex.org/W7160871604","doi":"https://doi.org/10.48550/arxiv.2605.07820","title":"Scaling Categorical Flow Maps","display_name":"Scaling Categorical Flow Maps","publication_year":2026,"publication_date":"2026-05-08","ids":{"openalex":"https://openalex.org/W7160871604","doi":"https://doi.org/10.48550/arxiv.2605.07820"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.07820","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.07820","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.07820","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135844255","display_name":"Oscar Davis","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Davis, Oscar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135907289","display_name":"Anastasiia Filippova","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Filippova, Anastasiia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135846452","display_name":"Pierre Ablin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ablin, Pierre","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120789496","display_name":"Victor Turrisi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Turrisi, Victor","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135859044","display_name":"Amitis Shidani","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shidani, Amitis","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017263427","display_name":"Marco Cuturi","orcid":"https://orcid.org/0000-0002-1934-0588"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cuturi, Marco","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135862519","display_name":"Louis B\u00e9thune","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"B\u00e9thune, Louis","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5135844255"],"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/T10028","display_name":"Topic Modeling","score":0.2800999879837036,"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"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.2800999879837036,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.10660000145435333,"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"}},{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.066600002348423,"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/categorical-variable","display_name":"Categorical variable","score":0.5807999968528748},{"id":"https://openalex.org/keywords/flow","display_name":"Flow (mathematics)","score":0.4997999966144562},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.4961000084877014},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.47290000319480896},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.4693000018596649},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.4429999887943268},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.4255000054836273},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.4189999997615814},{"id":"https://openalex.org/keywords/constraint","display_name":"Constraint (computer-aided design)","score":0.39250001311302185}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6269999742507935},{"id":"https://openalex.org/C5274069","wikidata":"https://www.wikidata.org/wiki/Q2285707","display_name":"Categorical variable","level":2,"score":0.5807999968528748},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.4997999966144562},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.4961000084877014},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.47290000319480896},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.4693000018596649},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.4496999979019165},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.4429999887943268},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.42559999227523804},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.4255000054836273},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.4189999997615814},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.39250001311302185},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.38589999079704285},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.3743000030517578},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.3564999997615814},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.3529999852180481},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.3375000059604645},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.3334999978542328},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3237000107765198},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.322299987077713},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3158999979496002},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.29109999537467957},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.2782000005245209},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.2777999937534332},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.27090001106262207},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.25189998745918274},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.07820","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.07820","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.07820","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.07820","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"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":{"Continuous":[0],"diffusion":[1,176],"and":[2,33,56,118,153,193,201],"flow":[3,50,113],"matching":[4,51],"models":[5,190],"could":[6],"represent":[7],"a":[8,21,48,54,110,122,144],"powerful":[9],"alternative":[10],"to":[11,160],"autoregressive":[12],"approaches":[13],"for":[14,27,147],"language":[15],"modelling":[16],"(LM),":[17],"as":[18,130,132,174],"they":[19,156],"unlock":[20],"host":[22],"of":[23,42,68,100,182],"advantages":[24],"currently":[25],"reserved":[26],"continuous":[28],"modalities,":[29],"including":[30],"accelerated":[31,69],"sampling":[32,70],"tilting.":[34],"Recently,":[35],"several":[36],"works":[37],"have":[38,63],"demonstrated":[39],"the":[40,57,66,82,98,150,162,171,183],"possibility":[41],"generating":[43],"discrete":[44,175],"data":[45,60],"continuously":[46],"by":[47],"simple":[49],"process":[52],"between":[53],"Gaussian":[55],"one-hot":[58],"encoded":[59],"distribution.":[61],"They":[62],"further":[64],"shown":[65],"feasibility":[67],"via":[71],"Categorical":[72],"Flow":[73],"Maps":[74],"(CFMs),":[75],"resulting":[76],"in":[77,81,129,149,170],"competitive":[78],"sample":[79],"quality":[80],"few-step":[83],"regime.":[84],"However,":[85],"this":[86,106],"method":[87],"had":[88],"only":[89],"been":[90],"evaluated":[91],"at":[92,191],"relatively":[93],"modest":[94],"scales":[95],"($<1$B),":[96],"leaving":[97],"question":[99],"its":[101],"scalability":[102],"completely":[103],"open.":[104],"In":[105],"article,":[107],"we":[108,142,179,194],"train":[109],"$1.7$B-parameter":[111],"base":[112],"model":[114,163],"on":[115,164,198],"$2.1$T":[116],"tokens":[117],"self-distill":[119],"it":[120],"into":[121],"CFM":[123],"that":[124,155,185],"generates":[125],"diverse,":[126],"high-quality":[127],"text":[128],"few":[131],"$4$":[133],"inference":[134],"steps":[135],"while":[136],"maintaining":[137],"near-data-level":[138],"token":[139],"entropy.":[140],"Furthermore,":[141],"introduce":[143],"likelihood":[145],"bound":[146],"CFMs":[148],"semi-discrete":[151],"setting,":[152],"show":[154],"can":[157],"be":[158],"used":[159],"score":[161],"standard":[165],"LM":[166],"benchmarks,":[167],"achieving":[168],"results":[169],"same":[172],"range":[173],"methods.":[177],"Finally,":[178],"uncover":[180],"some":[181],"challenges":[184],"arise":[186],"from":[187],"training":[188],"these":[189],"scale,":[192],"provide":[195],"prescriptive":[196],"insights":[197],"loss":[199],"weighting":[200],"time":[202],"scheduling.":[203]},"counts_by_year":[],"updated_date":"2026-05-12T06:14:25.881160","created_date":"2026-05-12T00:00:00"}