{"id":"https://openalex.org/W4415538849","doi":"https://doi.org/10.1145/3746027.3755160","title":"Less is More: High-value Data Selection for Visual Instruction Tuning","display_name":"Less is More: High-value Data Selection for Visual Instruction Tuning","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415538849","doi":"https://doi.org/10.1145/3746027.3755160"},"language":null,"primary_location":{"id":"doi:10.1145/3746027.3755160","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3755160","pdf_url":null,"source":null,"license":null,"license_id":null,"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":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119269640","display_name":"Zikang Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Zikang Liu","raw_affiliation_strings":["Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063459528","display_name":"Kun Zhou","orcid":"https://orcid.org/0000-0003-0650-9521"},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kun Zhou","raw_affiliation_strings":["School of Information, Renmin University of China, Beijing, China"],"affiliations":[{"raw_affiliation_string":"School of Information, Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037145565","display_name":"Wayne Xin Zhao","orcid":"https://orcid.org/0000-0002-8333-6196"},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wayne Xin Zhao","raw_affiliation_strings":["Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103243413","display_name":"Dawei Gao","orcid":"https://orcid.org/0009-0007-3882-5189"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dawei Gao","raw_affiliation_strings":["Alibaba Group, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Alibaba Group, Beijing, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046576694","display_name":"Yaliang Li","orcid":"https://orcid.org/0000-0002-4204-6096"},"institutions":[{"id":"https://openalex.org/I4210108985","display_name":"Bellevue Hospital Center","ror":"https://ror.org/01ky34z31","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I1283621791","https://openalex.org/I4210086933","https://openalex.org/I4210108985"]},{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yaliang Li","raw_affiliation_strings":["Alibaba Group, Bellevue, Washington, USA"],"affiliations":[{"raw_affiliation_string":"Alibaba Group, Bellevue, Washington, USA","institution_ids":["https://openalex.org/I4210095624","https://openalex.org/I4210108985"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5025631695","display_name":"Ji-Rong Wen","orcid":"https://orcid.org/0000-0002-9777-9676"},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ji-Rong Wen","raw_affiliation_strings":["Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5119269640"],"corresponding_institution_ids":["https://openalex.org/I78988378"],"apc_list":null,"apc_paid":null,"fwci":4.284,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.9497283,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"3712","last_page":"3721"},"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/T11439","display_name":"Video Analysis and Summarization","score":0.9959999918937683,"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":0.9958000183105469,"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/leverage","display_name":"Leverage (statistics)","score":0.6840000152587891},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.5976999998092651},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.45339998602867126},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.390500009059906},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.3711000084877014},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.34599998593330383},{"id":"https://openalex.org/keywords/heuristic","display_name":"Heuristic","score":0.3416999876499176}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8044000267982483},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.6840000152587891},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.5976999998092651},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5669000148773193},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5149000287055969},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.45339998602867126},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.390500009059906},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3711000084877014},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.34599998593330383},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.3416999876499176},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.3059000074863434},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.29249998927116394},{"id":"https://openalex.org/C172367668","wikidata":"https://www.wikidata.org/wiki/Q6504956","display_name":"Data visualization","level":3,"score":0.2718000113964081},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.26010000705718994},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.25760000944137573}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746027.3755160","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3755160","pdf_url":null,"source":null,"license":null,"license_id":null,"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":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W2251512949","https://openalex.org/W2560730294","https://openalex.org/W2886641317","https://openalex.org/W2947312908","https://openalex.org/W2963109634","https://openalex.org/W2979473749","https://openalex.org/W4312846625","https://openalex.org/W4367367040","https://openalex.org/W4367628410","https://openalex.org/W4376988980","https://openalex.org/W4383987918","https://openalex.org/W4385473806","https://openalex.org/W4386148496","https://openalex.org/W4389523832","https://openalex.org/W4392011716","https://openalex.org/W4393300611","https://openalex.org/W4401042981","https://openalex.org/W4402716477","https://openalex.org/W4402727764","https://openalex.org/W4403791888","https://openalex.org/W4404575065","https://openalex.org/W4404783431"],"related_works":[],"abstract_inverted_index":{"Visual":[0],"instruction":[1,32,90,110,153],"tuning":[2],"is":[3],"the":[4,17,53,71,82,88,94,108,114,121,127,136,147,151,181,194],"key":[5],"to":[6,104,145,180],"building":[7],"large":[8],"vision":[9],"language":[10],"models":[11],"(LVLMs),":[12],"which":[13,46,66],"can":[14,176],"greatly":[15,69],"improve":[16],"task":[18,134,137,148],"solving":[19],"and":[20,51,112,135,155,198],"generalization":[21],"capabilities.":[22],"Previous":[23],"work":[24],"mostly":[25],"collects":[26],"a":[27,43,61,98],"mixture":[28],"of":[29,63,73,193],"existing":[30],"visual":[31,89,109,152],"datasets":[33],"via":[34],"heuristic":[35],"ways":[36],"for":[37,159],"train-":[38],"ing":[39],"(even":[40],"more":[41],"than":[42],"million":[44],"instructions),":[45],"may":[47],"introduce":[48],"data":[49,100,111,175,199],"redundancy":[50,86,106],"increase":[52],"training":[54,115],"cost.":[55,116],"To":[56],"investigate":[57],"it,":[58],"we":[59,96,125,141],"conduct":[60],"series":[62],"empirical":[64],"studies,":[65],"show":[67,167],"that":[68,168],"reducing":[70],"amount":[72],"instructions":[74],"from":[75],"several":[76],"tasks":[77],"even":[78,188],"do":[79],"not":[80],"affect":[81],"performance,":[83],"indicating":[84],"significant":[85],"within":[87,107,150],"datasets.":[91],"Based":[92],"on":[93,120,131,164,191],"findings,":[95],"propose":[97],"high-value":[99,157],"selection":[101],"approach":[102,170],"TIVE,":[103,118],"eliminate":[105],"reduce":[113],"In":[117],"based":[119],"gradient-based":[122],"influence":[123,129],"functions,":[124],"estimate":[126],"instance":[128],"score":[130],"its":[132],"corresponding":[133],"difficulty":[138],"score.":[139],"Then,":[140],"leverage":[142],"these":[143],"scores":[144],"determine":[146],"proportion":[149],"subset,":[154],"select":[156],"instances":[158],"each":[160],"task,":[161],"respectively.":[162],"Experiments":[163],"various":[165],"LVLMs":[166],"our":[169],"using":[171],"only":[172],"about":[173],"15%":[174],"achieve":[177],"comparable":[178],"performance":[179],"full-data":[182],"fine-tuned":[183],"model":[184],"across":[185],"eight":[186],"benchmarks,":[187],"surpassing":[189],"it":[190],"four":[192],"benchmarks.":[195],"Our":[196],"code":[197],"will":[200],"be":[201],"publicly":[202],"released.":[203]},"counts_by_year":[{"year":2025,"cited_by_count":3}],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-25T00:00:00"}