{"id":"https://openalex.org/W7168967114","doi":"https://doi.org/10.48550/arxiv.2607.13661","title":"Fine-grained CLIP fine-tuning with self-annotated region alignment","display_name":"Fine-grained CLIP fine-tuning with self-annotated region alignment","publication_year":2026,"publication_date":"2026-07-15","ids":{"openalex":"https://openalex.org/W7168967114","doi":"https://doi.org/10.48550/arxiv.2607.13661"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.13661","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13661","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,"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.2607.13661","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5141019974","display_name":"Chenyang Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Chenyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141043385","display_name":"Wei Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140970453","display_name":"Antoni B. Chan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chan, Antoni B.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136748432","display_name":"Janet H. Hsiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hsiao, Janet H.","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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.7840999960899353,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.7840999960899353,"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":0.07500000298023224,"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":0.03180000185966492,"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/feature","display_name":"Feature (linguistics)","score":0.6690000295639038},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.5848000049591064},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.579800009727478},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5651000142097473},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5503000020980835},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.43459999561309814},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.423799991607666},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.38499999046325684}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7402999997138977},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.6690000295639038},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.602400004863739},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5848000049591064},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.579800009727478},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5651000142097473},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5503000020980835},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.43459999561309814},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4334999918937683},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.423799991607666},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.38499999046325684},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3822999894618988},{"id":"https://openalex.org/C2781122975","wikidata":"https://www.wikidata.org/wiki/Q16928266","display_name":"Semantic feature","level":2,"score":0.366100013256073},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.36550000309944153},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.33559998869895935},{"id":"https://openalex.org/C2983787585","wikidata":"https://www.wikidata.org/wiki/Q93586","display_name":"Feature matching","level":3,"score":0.30640000104904175},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.28780001401901245},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.25270000100135803},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2524999976158142},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.2506999969482422}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.13661","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13661","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,"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.2607.13661","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13661","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,"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":[{"id":"https://metadata.un.org/sdg/4","score":0.5744112133979797,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Contrastive":[0],"Language-Image":[1],"Pre-training":[2],"(CLIP)":[3],"has":[4],"been":[5],"shown":[6],"to":[7,17,26,49,81,87,91,126,176],"have":[8],"limitations":[9],"in":[10,37,132],"its":[11,18],"fine-grained":[12,52,129,181],"dense":[13,182],"feature":[14,183],"representation,":[15,184],"due":[16,80],"pre-training":[19,38],"focusing":[20],"on":[21,180,195],"matching":[22],"the":[23,31,51,77,82,93,128,133,138,155,189,192],"whole":[24],"image":[25],"a":[27,39,44,57,65,88,143],"text":[28],"description.":[29],"Considering":[30],"large":[32,89],"data":[33],"and":[34,85,96,158],"computational":[35],"burden":[36],"vision-language":[40],"model":[41],"from":[42,64,154],"scratch,":[43],"series":[45],"of":[46,54,67,191],"works":[47,62],"aim":[48],"enhance":[50],"ability":[53,102,131],"CLIP":[55,134,194],"through":[56],"fine-tuning":[58],"scheme.":[59],"However,":[60],"existing":[61],"suffer":[63],"variety":[66],"limitations:":[68],"additional":[69],"region":[70],"annotations":[71],"are":[72],"usually":[73,98],"required,":[74],"which":[75,119,150],"limits":[76],"semantic":[78],"diversity":[79],"predefined":[83],"categories":[84],"leads":[86,175],"effort":[90],"process":[92],"training":[94],"data;":[95],"they":[97],"sacrifice":[99],"CLIP's":[100],"original":[101,193],"for":[103,117],"global":[104,139],"visual":[105],"representation.":[106],"To":[107],"bypass":[108],"these":[109],"limitations,":[110],"we":[111],"propose":[112],"SFF-CLIP":[113,174],"(Self-annotated":[114],"Fine-grained":[115],"Fine-tuning":[116],"CLIP),":[118],"only":[120],"uses":[121],"image-text":[122],"pairs":[123],"as":[124,185,187],"input":[125,156],"boost":[127],"representation":[130],"fine-tuning,":[135],"while":[136],"maintaining":[137,188],"visual-semantic":[140],"consistency.":[141],"Concretely,":[142],"run-time":[144],"region-phrase":[145],"alignment":[146],"scheme":[147],"is":[148],"designed,":[149],"obtains":[151],"concept":[152],"phrases":[153],"sentence,":[157],"aligns":[159],"them":[160],"with":[161],"corresponding":[162],"extracted":[163],"region-based":[164],"features":[165],"using":[166],"text-specific":[167],"heat":[168],"maps.":[169],"Extensive":[170],"experiments":[171],"demonstrate":[172],"that":[173],"significant":[177],"performance":[178,190],"improvements":[179],"well":[186],"image-level":[196],"tasks.":[197],"Code":[198],"will":[199],"be":[200],"released":[201],"later.":[202]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-17T00:00:00"}