{"id":"https://openalex.org/W7166557614","doi":"https://doi.org/10.48550/arxiv.2606.27617","title":"Masked Language Flow Models","display_name":"Masked Language Flow Models","publication_year":2026,"publication_date":"2026-06-26","ids":{"openalex":"https://openalex.org/W7166557614","doi":"https://doi.org/10.48550/arxiv.2606.27617"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.27617","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27617","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.27617","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136158794","display_name":"Iskander Azangulov","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Azangulov, Iskander","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109818828","display_name":"Kianoosh Ashouritaklimi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ashouritaklimi, Kianoosh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139553279","display_name":"Leo Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Leo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026894129","display_name":"Simon Vary","orcid":"https://orcid.org/0000-0002-9929-4357"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vary, Simon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5091668854","display_name":"Patrick Rebeschini","orcid":"https://orcid.org/0000-0001-7772-4160"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rebeschini, Patrick","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/T10028","display_name":"Topic Modeling","score":0.41110000014305115,"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":0.41110000014305115,"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":0.14659999310970306,"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/T13629","display_name":"Text Readability and Simplification","score":0.05339999869465828,"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/security-token","display_name":"Security token","score":0.7788000106811523},{"id":"https://openalex.org/keywords/masking","display_name":"Masking (illustration)","score":0.583299994468689},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.5159000158309937},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5077999830245972},{"id":"https://openalex.org/keywords/flow","display_name":"Flow (mathematics)","score":0.4977000057697296},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.40369999408721924},{"id":"https://openalex.org/keywords/obstacle","display_name":"Obstacle","score":0.319599986076355}],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7788000106811523},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7470999956130981},{"id":"https://openalex.org/C2777402240","wikidata":"https://www.wikidata.org/wiki/Q6783436","display_name":"Masking (illustration)","level":2,"score":0.583299994468689},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5159000158309937},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5077999830245972},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.4977000057697296},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43849998712539673},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.40369999408721924},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3977999985218048},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3239000141620636},{"id":"https://openalex.org/C2776650193","wikidata":"https://www.wikidata.org/wiki/Q264661","display_name":"Obstacle","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.3158000111579895},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.31279999017715454},{"id":"https://openalex.org/C2985264121","wikidata":"https://www.wikidata.org/wiki/Q216320","display_name":"Continuous flow","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.30239999294281006},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.28690001368522644},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2815000116825104},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.26010000705718994},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2590999901294708},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.25839999318122864},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.2542000114917755},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.27617","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27617","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.27617","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27617","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":"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":{"Masked":[0,97],"Diffusion":[1],"Models":[2,40,100],"(MDMs)":[3],"promise":[4],"fast,":[5],"parallel":[6,29],"language":[7,181],"generation,":[8],"but":[9],"their":[10],"reverse":[11],"transition":[12],"factorises":[13],"across":[14],"token":[15,89],"positions":[16],"--":[17],"an":[18],"approximation":[19],"that":[20,50,64,149,179],"breaks":[21],"down":[22],"in":[23,57],"the":[24,34,154,176],"few-step":[25],"sampling":[26],"regime":[27],"where":[28],"generation":[30,123],"ought":[31],"to":[32,86,112,131,160,186],"provide":[33],"greatest":[35],"efficiency":[36],"gains.":[37],"Flow":[38,99],"Language":[39,98],"(FLMs)":[41],"sidestep":[42],"this":[43,72,142],"limitation":[44],"by":[45],"learning":[46],"a":[47,61,108,137,146],"continuous":[48,109,125,151],"flow":[49,62],"transports":[51],"noise":[52],"toward":[53],"clean":[54,117],"sequences":[55],"represented":[56],"Euclidean":[58],"space,":[59],"inducing":[60],"map":[63],"can":[65,183],"be":[66,132,184],"distilled":[67],"for":[68,80,175],"single-step":[69],"generation.":[70,91],"However,":[71],"makes":[73],"complex":[74],"tasks":[75],"requiring":[76],"multi-step":[77,163],"reasoning":[78,189],"problematic":[79],"FLMs,":[81],"as":[82],"FLMs":[83,106],"are":[84],"forced":[85],"decode":[87],"every":[88],"during":[90],"To":[92],"address":[93],"this,":[94],"we":[95,144],"introduce":[96],"(MLFMs),":[101],"which":[102],"incorporate":[103],"masking":[104],"into":[105,134],"using":[107],"stochastic":[110],"interpolant":[111],"bridge":[113],"partially":[114],"masked":[115],"and":[116,127,171,173,190],"sequences.":[118],"This":[119],"design":[120],"enables":[121],"conditional":[122],"via":[124],"flows":[126],"allows":[128],"pretrained":[129],"MDMs":[130],"converted":[133],"MLFMs":[135],"through":[136],"simple,":[138],"lightweight":[139],"adaptation.":[140],"Leveraging":[141],"flexibility,":[143],"propose":[145],"novel":[147],"sampler":[148],"alternates":[150],"denoising":[152],"with":[153],"discrete":[155],"unmasking":[156],"of":[157],"confident":[158],"tokens":[159],"better":[161],"support":[162],"reasoning.":[164],"We":[165],"evaluate":[166],"our":[167],"approach":[168],"on":[169],"GSM8K":[170],"MT-Bench":[172],"find,":[174],"first":[177],"time,":[178],"flow-based":[180],"models":[182],"scaled":[185],"solve":[187],"downstream":[188],"instruction-following":[191],"tasks.":[192]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-30T00:00:00"}