{"id":"https://openalex.org/W4415540548","doi":"https://doi.org/10.1145/3746027.3754714","title":"SynC: Synthetic Image Caption Dataset Refinement with One-to-many Mapping for Zero-shot Image Captioning","display_name":"SynC: Synthetic Image Caption Dataset Refinement with One-to-many Mapping for Zero-shot Image Captioning","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415540548","doi":"https://doi.org/10.1145/3746027.3754714"},"language":null,"primary_location":{"id":"doi:10.1145/3746027.3754714","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3754714","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3746027.3754714","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5108808046","display_name":"Sun I. Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Si-Woo Kim","raw_affiliation_strings":["Hanyang University, Seoul, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108943847","display_name":"MinJu Jeon","orcid":null},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"MinJu Jeon","raw_affiliation_strings":["Hanyang University, Seoul, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064180074","display_name":"Ye-Chan Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Ye-Chan Kim","raw_affiliation_strings":["Hanyang University, Seoul, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101530945","display_name":"Soeun Lee","orcid":"https://orcid.org/0000-0002-0716-5962"},"institutions":[{"id":"https://openalex.org/I4210089867","display_name":"Korea Petroleum Group (South Korea)","ror":"https://ror.org/007mm3r71","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210089867"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Soeun Lee","raw_affiliation_strings":["AI R&D Division, CJ Group, Seoul, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"AI R&D Division, CJ Group, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4210089867"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101736514","display_name":"Taewhan Kim","orcid":"https://orcid.org/0000-0002-2126-0704"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Taewhan Kim","raw_affiliation_strings":["Hanyang University, Seoul, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100393157","display_name":"Dong-Jin Kim","orcid":"https://orcid.org/0000-0001-7231-7494"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Dong-Jin Kim","raw_affiliation_strings":["Hanyang University, Seoul, Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Hanyang University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4575257"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5108808046"],"corresponding_institution_ids":["https://openalex.org/I4575257"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.32643999,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2683","last_page":"2692"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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":1.0,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9966999888420105,"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":0.9879999756813049,"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/closed-captioning","display_name":"Closed captioning","score":0.9050999879837036},{"id":"https://openalex.org/keywords/sync","display_name":"sync","score":0.7734000086784363},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.6579999923706055},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6482999920845032},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.6362000107765198},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.37770000100135803}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.9050999879837036},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8410000205039978},{"id":"https://openalex.org/C3913047","wikidata":"https://www.wikidata.org/wiki/Q1956265","display_name":"sync","level":3,"score":0.7734000086784363},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.6579999923706055},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6482999920845032},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.6362000107765198},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6202999949455261},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4153999984264374},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.37770000100135803},{"id":"https://openalex.org/C2989087649","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Image synthesis","level":3,"score":0.35370001196861267},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.34950000047683716},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3073999881744385},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2872999906539917},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2630000114440918},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.2506999969482422},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746027.3754714","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3754714","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3746027.3754714","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3754714","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W1905882502","https://openalex.org/W1956340063","https://openalex.org/W2185175083","https://openalex.org/W2886641317","https://openalex.org/W2904565150","https://openalex.org/W2963743213","https://openalex.org/W2979826702","https://openalex.org/W3176641147","https://openalex.org/W4312933868","https://openalex.org/W4386148430","https://openalex.org/W4390873312","https://openalex.org/W4400352269","https://openalex.org/W4402727764"],"related_works":[],"abstract_inverted_index":{"Zero-shot":[0],"Image":[1],"Captioning":[2],"(ZIC)":[3],"increasingly":[4],"utilizes":[5],"synthetic":[6,45,77,104,130,208],"datasets":[7,106],"generated":[8],"by":[9,140,162],"text-to-image":[10],"(T2I)":[11],"models":[12,24,187],"to":[13,102,120,166,210],"mitigate":[14],"the":[15,73,121,129,159,168],"need":[16],"for":[17,60,72,107,147,205],"costly":[18],"manual":[19],"annotation.":[20],"However,":[21,67],"these":[22,68],"T2I":[23],"often":[25],"produce":[26],"images":[27,85,125,146],"that":[28,48,157,177],"exhibit":[29],"semantic":[30],"misalignments":[31],"with":[32],"their":[33],"corresponding":[34],"input":[35],"captions":[36,80,119],"(e.g.,":[37],"missing":[38],"objects,":[39],"incorrect":[40],"attributes),":[41],"resulting":[42],"in":[43,64,197],"noisy":[44,62],"image-caption":[46,105],"pairs":[47],"can":[49],"hinder":[50],"model":[51],"training.":[52],"Existing":[53],"dataset":[54],"pruning":[55],"techniques":[56],"are":[57,70,81],"largely":[58],"designed":[59,101],"removing":[61],"text":[63],"web-crawled":[65],"data.":[66],"methods":[69],"ill-suited":[71],"distinct":[74],"challenges":[75],"of":[76,110],"data,":[78],"where":[79],"typically":[82],"well-formed,":[83],"but":[84],"may":[86],"be":[87],"inaccurate":[88],"representations.":[89],"To":[90],"address":[91],"this":[92],"gap,":[93],"we":[94],"introduce":[95],"SynC,":[96],"a":[97,136,153],"novel":[98],"framework":[99],"specifically":[100],"refine":[103],"ZIC.":[108,212],"Instead":[109],"conventional":[111],"filtering":[112],"or":[113],"regeneration,":[114],"SynC":[115,178,200],"focuses":[116],"on":[117,188],"reassigning":[118],"most":[122],"semantically":[123],"aligned":[124],"already":[126],"present":[127],"within":[128],"image":[131,161],"pool.":[132],"Our":[133],"approach":[134],"employs":[135],"one-to-many":[137],"mapping":[138],"strategy":[139,204],"initially":[141],"retrieving":[142],"multiple":[143],"relevant":[144],"candidate":[145],"each":[148],"caption.":[149],"We":[150],"then":[151],"apply":[152],"cycle-consistency-inspired":[154],"alignment":[155],"scorer":[156],"selects":[158],"best":[160],"verifying":[163],"its":[164],"ability":[165],"retrieve":[167],"original":[169],"caption":[170],"via":[171],"image-to-text":[172],"retrieval.":[173],"Extensive":[174],"evaluations":[175],"demonstrate":[176],"consistently":[179],"and":[180],"significantly":[181],"improves":[182],"performance":[183],"across":[184],"various":[185],"ZIC":[186],"standard":[189],"benchmarks":[190],"(MS-COCO,":[191],"Flickr30k,":[192],"NoCaps),":[193],"achieving":[194],"state-of-the-art":[195],"results":[196],"several":[198],"scenarios.":[199],"offers":[201],"an":[202],"effective":[203],"curating":[206],"refined":[207],"data":[209],"enhance":[211]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-25T00:00:00"}