{"id":"https://openalex.org/W4392617203","doi":"https://doi.org/10.1145/3774904.3792371","title":"Adaptive Task Balancing for Visual Instruction Tuning via Inter-Task Contribution and Intra-Task Difficulty","display_name":"Adaptive Task Balancing for Visual Instruction Tuning via Inter-Task Contribution and Intra-Task Difficulty","publication_year":2026,"publication_date":"2026-04-09","ids":{"openalex":"https://openalex.org/W4392617203","doi":"https://doi.org/10.1145/3774904.3792371"},"language":"en","primary_location":{"id":"doi:10.1145/3774904.3792371","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774904.3792371","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 ACM Web Conference 2026","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3774904.3792371","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Yanqi Dai","orcid":"https://orcid.org/0009-0000-5441-3263"},"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":"Yanqi Dai","raw_affiliation_strings":["Renmin University of China, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0000-5441-3263","affiliations":[{"raw_affiliation_string":"Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yong Wang","orcid":"https://orcid.org/0009-0008-7878-493X"},"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":"Yong Wang","raw_affiliation_strings":["AMAP, Alibaba Group, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0008-7878-493X","affiliations":[{"raw_affiliation_string":"AMAP, Alibaba Group, Beijing, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zebin You","orcid":"https://orcid.org/0009-0005-3577-8693"},"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":"Zebin You","raw_affiliation_strings":["Renmin University of China, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0005-3577-8693","affiliations":[{"raw_affiliation_string":"Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Dong Jing","orcid":"https://orcid.org/0009-0008-0769-8102"},"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":"Dong Jing","raw_affiliation_strings":["Renmin University of China, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0008-0769-8102","affiliations":[{"raw_affiliation_string":"Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Xiangxiang Chu","orcid":"https://orcid.org/0000-0003-2548-0605"},"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":"Xiangxiang Chu","raw_affiliation_strings":["AMAP, Alibaba Group, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2548-0605","affiliations":[{"raw_affiliation_string":"AMAP, Alibaba Group, Beijing, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"last","author":{"id":null,"display_name":"Zhiwu Lu","orcid":"https://orcid.org/0000-0001-6429-7956"},"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":"Zhiwu Lu","raw_affiliation_strings":["Renmin University of China, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-6429-7956","affiliations":[{"raw_affiliation_string":"Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I78988378"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00072025,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3893","last_page":"3904"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11707","display_name":"Gaze Tracking and Assistive Technology","score":0.8919000029563904,"subfield":{"id":"https://openalex.org/subfields/1709","display_name":"Human-Computer Interaction"},"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/T11707","display_name":"Gaze Tracking and Assistive Technology","score":0.8919000029563904,"subfield":{"id":"https://openalex.org/subfields/1709","display_name":"Human-Computer Interaction"},"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/T11122","display_name":"Online Learning and Analytics","score":0.8891000151634216,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/T11283","display_name":"Experimental Learning in Engineering","score":0.8238000273704529,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.7588415145874023},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6839486956596375},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.43700093030929565},{"id":"https://openalex.org/keywords/cognitive-psychology","display_name":"Cognitive psychology","score":0.3344007134437561},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.22605803608894348},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.06188863515853882}],"concepts":[{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.7588415145874023},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6839486956596375},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.43700093030929565},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.3344007134437561},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.22605803608894348},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.06188863515853882},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3774904.3792371","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774904.3792371","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 ACM Web Conference 2026","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2403.04343","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2403.04343","pdf_url":"https://arxiv.org/pdf/2403.04343","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2403.04343","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2403.04343","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.1145/3774904.3792371","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774904.3792371","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 ACM Web Conference 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2382290278","https://openalex.org/W2478288626","https://openalex.org/W4391913857","https://openalex.org/W3196817267","https://openalex.org/W4211075255"],"abstract_inverted_index":{"Visual":[0],"instruction":[1,51,178],"tuning":[2,52],"is":[3],"a":[4,42,99],"key":[5],"training":[6],"stage":[7],"of":[8,98,108,155],"large":[9],"multimodal":[10],"models.":[11],"However,":[12],"when":[13],"learning":[14,74,96,132],"multiple":[15],"visual":[16,50,61,177],"tasks":[17,109],"simultaneously,":[18],"this":[19,38],"approach":[20,47,167],"often":[21],"results":[22],"in":[23,176],"suboptimal":[24],"and":[25,88,127,142,149,171,183],"imbalanced":[26],"overall":[27,147,174],"performance":[28,79,156,175],"due":[29],"to":[30,83,118],"latent":[31],"knowledge":[32,85],"conflicts":[33],"across":[34,86],"tasks.":[35],"To":[36],"mitigate":[37],"issue,":[39],"we":[40,55,103],"propose":[41,104],"novel":[43],"Adaptive":[44],"Task":[45],"Balancing":[46],"tailored":[48],"for":[49,60],"(VisATB).":[53],"Specifically,":[54],"measure":[56],"two":[57],"critical":[58],"dimensions":[59],"task":[62,76,137],"balancing":[63],"based":[64],"on":[65,80,145,160],"validation":[66],"performance:":[67],"(1)":[68],"Inter-Task":[69],"Contribution,":[70],"the":[71,78,94,140,150,153],"mechanism":[72],"where":[73],"one":[75],"enhances":[77],"others":[81],"owing":[82],"shared":[84],"tasks,":[87],"(2)":[89],"Intra-Task":[90],"Difficulty,":[91],"which":[92],"denotes":[93],"inherent":[95],"difficulty":[97],"single":[100],"task.":[101],"Furthermore,":[102],"prioritizing":[105],"three":[106,136,161],"categories":[107],"with":[110],"greater":[111],"weight:":[112],"those":[113,120,128],"that":[114,121,129,164],"offer":[115],"substantial":[116],"contributions":[117,124],"others,":[119,126],"receive":[122],"minimal":[123],"from":[125],"present":[130],"high":[131],"difficulties.":[133],"Among":[134],"these":[135],"weighting":[138],"strategies,":[139],"first":[141],"third":[143],"focus":[144],"improving":[146],"performance,":[148],"second":[151],"targets":[152],"mitigation":[154],"imbalance.":[157],"Extensive":[158],"experiments":[159],"benchmarks":[162],"demonstrate":[163],"our":[165],"VisATB":[166],"consistently":[168],"achieves":[169],"superior":[170],"more":[172],"balanced":[173],"tuning.":[179],"The":[180],"data,":[181],"code,":[182],"models":[184],"are":[185],"available":[186],"at":[187],"https://github.com/YanqiDai/VisATB.":[188]},"counts_by_year":[],"updated_date":"2026-05-04T08:30:34.212998","created_date":"2024-03-09T00:00:00"}
