{"id":"https://openalex.org/W4415539179","doi":"https://doi.org/10.1145/3746027.3758222","title":"SynthVLM: Towards High-Quality and Efficient Synthesis of Image-Caption Datasets for Vision-Language Models","display_name":"SynthVLM: Towards High-Quality and Efficient Synthesis of Image-Caption Datasets for Vision-Language Models","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415539179","doi":"https://doi.org/10.1145/3746027.3758222"},"language":null,"primary_location":{"id":"doi:10.1145/3746027.3758222","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3758222","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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3746027.3758222","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5109886881","display_name":"Zheng Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng Liu","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0006-1277-0493","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Hao Liang","orcid":"https://orcid.org/0009-0000-2963-2210"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Liang","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0000-2963-2210","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062449586","display_name":"Bozhou Li","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bozhou Li","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0001-7519-5733","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Wentao Xiong","orcid":"https://orcid.org/0000-0002-9033-587X"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wentao Xiong","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-9033-587X","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025013542","display_name":"Chong Chen","orcid":"https://orcid.org/0000-0003-1417-2295"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chong Chen","raw_affiliation_strings":["Huawei Technologies Ltd., Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1417-2295","affiliations":[{"raw_affiliation_string":"Huawei Technologies Ltd., Beijing, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101615091","display_name":"Conghui He","orcid":"https://orcid.org/0000-0001-8697-695X"},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Conghui He","raw_affiliation_strings":["Shanghai AI Laboratory, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0001-8697-695X","affiliations":[{"raw_affiliation_string":"Shanghai AI Laboratory, Shanghai, China","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008772211","display_name":"Wentao Zhang","orcid":"https://orcid.org/0000-0002-7532-5550"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wentao Zhang","raw_affiliation_strings":["Center for Machine Learning Research, Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-7532-5550","affiliations":[{"raw_affiliation_string":"Center for Machine Learning Research, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5062357883","display_name":"Bin Cui","orcid":"https://orcid.org/0000-0003-1681-4677"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bin Cui","raw_affiliation_strings":["School of Computer Science & Key Lab of High Confidence Software Technologies (MOE), Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1681-4677","affiliations":[{"raw_affiliation_string":"School of Computer Science & Key Lab of High Confidence Software Technologies (MOE), Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.32777171,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"12807","last_page":"12814"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9965000152587891,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9951000213623047,"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/language-model","display_name":"Language model","score":0.47440001368522644},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.4618000090122223},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.42579999566078186},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.37860000133514404},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.3260999917984009},{"id":"https://openalex.org/keywords/web-application","display_name":"Web application","score":0.305400013923645},{"id":"https://openalex.org/keywords/noisy-data","display_name":"Noisy data","score":0.28040000796318054}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8296999931335449},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5741000175476074},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.47440001368522644},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.4618000090122223},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.42579999566078186},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4088999927043915},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4065000116825104},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.37860000133514404},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.33390000462532043},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3260999917984009},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3156000077724457},{"id":"https://openalex.org/C118643609","wikidata":"https://www.wikidata.org/wiki/Q189210","display_name":"Web application","level":2,"score":0.305400013923645},{"id":"https://openalex.org/C2781170535","wikidata":"https://www.wikidata.org/wiki/Q30587856","display_name":"Noisy data","level":2,"score":0.28040000796318054},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.275299996137619},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.2752000093460083},{"id":"https://openalex.org/C91632574","wikidata":"https://www.wikidata.org/wiki/Q15088675","display_name":"Data curation","level":2,"score":0.266400009393692},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.25699999928474426},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.25040000677108765},{"id":"https://openalex.org/C171686336","wikidata":"https://www.wikidata.org/wiki/Q3532085","display_name":"Topic model","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746027.3758222","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3758222","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:10.1145/3746027.3758222","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746027.3758222","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":[{"id":"https://openalex.org/G4954098831","display_name":null,"funder_award_id":"92470121,62402016","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8708956154","display_name":null,"funder_award_id":"2024YFA1014003","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":9,"referenced_works":["https://openalex.org/W2133665775","https://openalex.org/W4364382874","https://openalex.org/W4366823186","https://openalex.org/W4383472654","https://openalex.org/W4386128198","https://openalex.org/W4391032878","https://openalex.org/W4404724811","https://openalex.org/W4405616836","https://openalex.org/W4405853951"],"related_works":[],"abstract_inverted_index":{"Vision-Language":[0],"Models":[1],"(VLMs)":[2],"have":[3],"recently":[4],"emerged,":[5],"demonstrating":[6,155],"remarkable":[7],"vision-understanding":[8],"capabilities.":[9],"However,":[10],"training":[11],"these":[12],"models":[13,59,115,133],"requires":[14],"large-scale":[15],"datasets,":[16],"which":[17,120],"brings":[18],"challenges":[19],"related":[20],"to":[21,63],"efficiency,":[22],"effectiveness,":[23],"and":[24,39,60,65,88,95,118,146,166,171],"quality":[25],"of":[26,85,111],"web":[27],"data.":[28,143],"In":[29,92],"this":[30,104],"paper,":[31],"we":[32,106],"introduce":[33,79],"SynthVLM,":[34],"a":[35,81,108],"new":[36,109],"data":[37,169],"synthesis":[38],"curation":[40],"method":[41],"for":[42],"generating":[43,170],"image-caption":[44,90],"pairs.":[45,76,91],"Unlike":[46],"traditional":[47,100],"methods,":[48],"where":[49],"captions":[50,62],"are":[51],"generated":[52],"from":[53,68],"images,":[54],"SynthVLM":[55],"utilizes":[56],"advanced":[57],"diffusion":[58],"high-quality":[61,82,158],"synthesize":[64],"select":[66],"images":[67],"text":[69],"captions,":[70],"thereby":[71],"creating":[72],"precisely":[73],"aligned":[74],"image-text":[75],"We":[77],"further":[78],"SynthVLM-100K,":[80],"dataset":[83,160,165],"consisting":[84],"100K":[86],"curated":[87],"synthesized":[89],"both":[93],"model":[94],"human":[96],"evaluations,":[97],"SynthVLM-100K":[98,159],"outperforms":[99],"real-world":[101],"datasets.":[102],"Leveraging":[103],"dataset,":[105],"develop":[107],"family":[110],"multimodal":[112],"large":[113],"language":[114,162],"(MLLMs),":[116],"SynthVLM-7B":[117,145],"SynthVLM-13B,":[119],"achieve":[121],"state-of-the-art":[122],"(SOTA)":[123],"performance":[124,150],"on":[125,151],"various":[126],"vision":[127],"question-answering":[128],"(VQA)":[129],"tasks.":[130],"Notably,":[131],"our":[132],"outperform":[134],"LLaVA":[135],"across":[136],"most":[137],"metrics":[138],"with":[139],"only":[140],"18%":[141],"pretrain":[142],"Furthermore,":[144],"SynthVLM-13B":[147],"attain":[148],"SOTA":[149],"the":[152,157,167],"MMLU":[153],"benchmark,":[154],"that":[156],"preserves":[161],"abilities.":[163],"Our":[164],"complete":[168],"curating":[172],"methods":[173],"can":[174],"be":[175],"found":[176],"in":[177],"https://github.com/starriver030515/SynthVLM.":[178]},"counts_by_year":[],"updated_date":"2026-08-01T09:00:35.917206","created_date":"2025-10-25T00:00:00"}