{"id":"https://openalex.org/W7140314897","doi":"https://doi.org/10.48550/arxiv.2603.22364","title":"MCLR: Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives","display_name":"MCLR: Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives","publication_year":2026,"publication_date":"2026-03-23","ids":{"openalex":"https://openalex.org/W7140314897","doi":"https://doi.org/10.48550/arxiv.2603.22364"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.22364","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.22364","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.22364","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130609501","display_name":"Xiang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Li, Xiang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130574808","display_name":"Yixuan Jia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia, Yixuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130567919","display_name":"Xiao Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Xiao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027207271","display_name":"Jeffrey A. Fessler","orcid":"https://orcid.org/0000-0001-9998-3315"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fessler, Jeffrey A.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130565545","display_name":"Rongrong Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Rongrong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130578505","display_name":"Qing Qu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qu, Qing","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5130609501"],"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":0.8669999837875366,"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":0.8669999837875366,"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/T11448","display_name":"Face recognition and analysis","score":0.026399999856948853,"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":0.02459999918937683,"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/equivalence","display_name":"Equivalence (formal languages)","score":0.6481000185012817},{"id":"https://openalex.org/keywords/heuristic","display_name":"Heuristic","score":0.5489000082015991},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.5462999939918518},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.5026000142097473},{"id":"https://openalex.org/keywords/maximization","display_name":"Maximization","score":0.476500004529953},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.4560000002384186},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4275999963283539},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.3808000087738037}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6832000017166138},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.6481000185012817},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.5489000082015991},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.5462999939918518},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5415999889373779},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.5026000142097473},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4887999892234802},{"id":"https://openalex.org/C2776330181","wikidata":"https://www.wikidata.org/wiki/Q18358244","display_name":"Maximization","level":2,"score":0.476500004529953},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.4560000002384186},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4275999963283539},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3808000087738037},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.34529998898506165},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.34380000829696655},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.3154999911785126},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.31049999594688416},{"id":"https://openalex.org/C2983787585","wikidata":"https://www.wikidata.org/wiki/Q93586","display_name":"Feature matching","level":3,"score":0.2971999943256378},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2833000123500824},{"id":"https://openalex.org/C2985793214","wikidata":"https://www.wikidata.org/wiki/Q3274096","display_name":"Utility maximization","level":2,"score":0.2773999869823456},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.2669999897480011},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.25679999589920044},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.25540000200271606}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.22364","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.22364","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.22364","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.22364","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":"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":{"Diffusion":[0],"models":[1,32],"have":[2],"achieved":[3],"state-of-the-art":[4],"performance":[5],"in":[6,57,72],"generative":[7],"modeling,":[8],"but":[9],"their":[10],"success":[11],"often":[12],"relies":[13],"heavily":[14],"on":[15],"classifier-free":[16,168],"guidance":[17,54,169],"(CFG),":[18],"an":[19],"inference-time":[20,53,82,137],"heuristic":[21],"that":[22,77,112,148],"modifies":[23],"the":[24,44,49,65,149,154],"sampling":[25],"trajectory.":[26],"From":[27],"a":[28,73,95,108,144,158,164,174],"theoretical":[29,145],"perspective,":[30],"diffusion":[31,100],"trained":[33],"with":[34,121],"standard":[35,78,99,127],"denoising":[36],"score":[37,151],"matching":[38],"(DSM)":[39],"are":[40],"expected":[41],"to":[42,87,157],"recover":[43],"target":[45],"data":[46],"distribution,":[47],"raising":[48],"question":[50],"of":[51,98,177],"why":[52],"is":[55,152],"necessary":[56],"practice.":[58],"In":[59],"this":[60],"work,":[61],"we":[62,105,142],"ask":[63],"whether":[64],"DSM":[66],"training":[67],"objective":[68,111],"can":[69],"be":[70],"modified":[71],"principled":[74,109],"manner":[75],"such":[76],"reverse-time":[79],"sampling,":[80,128],"without":[81,135],"guidance,":[83],"yields":[84],"effects":[85],"comparable":[86,130],"CFG.":[88,178],"We":[89],"identify":[90],"insufficient":[91],"inter-class":[92,115],"separation":[93],"as":[94],"key":[96],"limitation":[97],"models.":[101],"To":[102],"address":[103],"this,":[104],"propose":[106],"MCLR,":[107],"alignment":[110],"explicitly":[113],"maximizes":[114],"likelihood-ratios":[116],"during":[117],"training.":[118],"Models":[119],"fine-tuned":[120],"MCLR":[122,160],"exhibit":[123],"CFG-like":[124],"improvements":[125],"under":[126],"achieving":[129],"qualitative":[131],"and":[132,170],"quantitative":[133],"gains":[134],"requiring":[136],"guidance.":[138],"Beyond":[139],"empirical":[140],"benefits,":[141],"provide":[143],"result":[146],"showing":[147],"CFG-guided":[150],"exactly":[153],"optimal":[155],"solution":[156],"weighted":[159],"objective.":[161],"This":[162],"establishes":[163],"formal":[165],"equivalence":[166],"between":[167],"alignment-based":[171],"objectives,":[172],"offering":[173],"mechanistic":[175],"interpretation":[176]},"counts_by_year":[],"updated_date":"2026-04-11T06:13:24.991567","created_date":"2026-03-26T00:00:00"}