{"id":"https://openalex.org/W7165666857","doi":"https://doi.org/10.48550/arxiv.2606.22550","title":"Training-Free Semantic Correction for Autoregressive Visual Models","display_name":"Training-Free Semantic Correction for Autoregressive Visual Models","publication_year":2026,"publication_date":"2026-06-21","ids":{"openalex":"https://openalex.org/W7165666857","doi":"https://doi.org/10.48550/arxiv.2606.22550"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.22550","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.22550","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.2606.22550","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139193650","display_name":"Junhao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Junhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139177467","display_name":"Chanyu Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Chanyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139194317","display_name":"Zheqi Lv","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lv, Zheqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120655297","display_name":"Keting Yin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yin, Keting","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139207514","display_name":"Shengyu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Shengyu","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":0.8888000249862671,"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":0.8888000249862671,"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":0.04899999871850014,"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":0.014999999664723873,"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/autoregressive-model","display_name":"Autoregressive model","score":0.630299985408783},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5306000113487244},{"id":"https://openalex.org/keywords/semantic-data-model","display_name":"Semantic data model","score":0.47749999165534973},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.4715000092983246},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.44190001487731934},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.4350999891757965},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.3955000042915344},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.37369999289512634}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8098000288009644},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.630299985408783},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6198999881744385},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5306000113487244},{"id":"https://openalex.org/C90312973","wikidata":"https://www.wikidata.org/wiki/Q7449052","display_name":"Semantic data model","level":2,"score":0.47749999165534973},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4715000092983246},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4453999996185303},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.44190001487731934},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.4350999891757965},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41909998655319214},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.3955000042915344},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.37369999289512634},{"id":"https://openalex.org/C534262118","wikidata":"https://www.wikidata.org/wiki/Q177719","display_name":"Medical diagnosis","level":2,"score":0.3736000061035156},{"id":"https://openalex.org/C55020928","wikidata":"https://www.wikidata.org/wiki/Q3813865","display_name":"Image quality","level":3,"score":0.3443000018596649},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3407999873161316},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3384000062942505},{"id":"https://openalex.org/C2775955345","wikidata":"https://www.wikidata.org/wiki/Q7449071","display_name":"Semantic mapping","level":2,"score":0.32690000534057617},{"id":"https://openalex.org/C197914299","wikidata":"https://www.wikidata.org/wiki/Q18650","display_name":"Semantic memory","level":3,"score":0.3167000114917755},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.271699994802475},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.26499998569488525},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.25189998745918274}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.22550","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.22550","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.2606.22550","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.22550","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":[{"id":"https://metadata.un.org/sdg/4","score":0.5372909307479858,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Autoregressive":[0],"visual":[1],"models":[2],"(AVMs)":[3],"based":[4],"on":[5,99,158],"next-scale":[6],"prediction":[7],"have":[8],"emerged":[9],"as":[10],"a":[11,105],"prominent":[12],"paradigm":[13],"for":[14,119],"image":[15,160],"and":[16,38,58,85,102,146,161,170],"video":[17,162],"synthesis.":[18],"However,":[19],"decomposing":[20],"the":[21,42,45,91,115,130,141,148,154],"generation":[22,67,79,149],"process":[23],"into":[24,56,90,114],"discrete":[25],"scales":[26],"with":[27,153],"varying":[28],"granularities":[29],"in":[30],"AVM":[31,52,66,116],"makes":[32],"semantic":[33,82,121,135,168],"errors":[34,83,136],"difficult":[35],"to":[36,50,64,88,151],"identify":[37],"correct,":[39],"thereby":[40],"undermining":[41],"quality":[43,68],"of":[44],"final":[46,92],"output.":[47,93],"Prior":[48],"efforts":[49,63],"enhance":[51,65],"can":[53],"be":[54],"categorized":[55],"training-based":[57,62],"training-free":[59,75,100],"approaches.":[60],"Although":[61],"come":[69],"at":[70],"substantial":[71],"computational":[72],"cost,":[73],"existing":[74],"methods":[76],"neglect":[77],"intermediate":[78,138],"states,":[80,139],"leaving":[81],"undiagnosed":[84],"allowing":[86],"them":[87],"accumulate":[89],"In":[94],"this":[95],"paper,":[96],"we":[97],"focus":[98],"paradigms":[101],"propose":[103],"Gazer,":[104],"framework":[106],"that":[107,165],"integrates":[108],"multimodal":[109],"large":[110],"language":[111],"model":[112],"feedback":[113],"sampling":[117],"loop":[118],"in-generation":[120],"correction.":[122],"Concretely,":[123],"Gazer":[124,166],"operates":[125],"via":[126],"two":[127],"cooperating":[128],"stages:":[129],"Reflective":[131],"Diagnosis":[132],"stage":[133,144],"diagnoses":[134],"from":[137],"while":[140],"Semantic":[142],"Correction":[143],"rewinds":[145],"rectifies":[147],"trajectory":[150],"realign":[152],"target":[155],"prompt.":[156],"Experiments":[157],"compositional":[159,171],"benchmarks":[163],"demonstrate":[164],"improves":[167],"alignment":[169],"accuracy":[172],"across":[173],"multiple":[174],"AVMs":[175],"without":[176],"additional":[177],"training.":[178]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-24T00:00:00"}