{"id":"https://openalex.org/W4406231070","doi":"https://doi.org/10.48550/arxiv.2501.04155","title":"MM-GEN: Enhancing Task Performance Through Targeted Multimodal Data Curation","display_name":"MM-GEN: Enhancing Task Performance Through Targeted Multimodal Data Curation","publication_year":2025,"publication_date":"2025-01-07","ids":{"openalex":"https://openalex.org/W4406231070","doi":"https://doi.org/10.48550/arxiv.2501.04155"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2501.04155","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2501.04155","pdf_url":"https://arxiv.org/pdf/2501.04155","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2501.04155","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5062849115","display_name":"Siddharth Joshi","orcid":"https://orcid.org/0000-0002-9201-9678","observed_orcids":["https://orcid.org/0000-0002-9201-9678"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Joshi, Siddharth","raw_affiliation_strings":["Microsoft Research","UCLA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]},{"raw_affiliation_string":"UCLA","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011998621","display_name":"Besmira Nushi","orcid":"https://orcid.org/0000-0002-7554-8586","observed_orcids":["https://orcid.org/0000-0002-7554-8586"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nushi, Besmira","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012491878","display_name":"Vidhisha Balachandran","orcid":"https://orcid.org/0009-0009-0465-0098","observed_orcids":["https://orcid.org/0009-0009-0465-0098"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Balachandran, Vidhisha","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009092703","display_name":"Varun Chandrasekaran","orcid":"https://orcid.org/0000-0003-0661-6378","observed_orcids":["https://orcid.org/0000-0003-0661-6378"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]},{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chandrasekaran, Varun","raw_affiliation_strings":["Microsoft Research","UIUC"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]},{"raw_affiliation_string":"UIUC","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045147286","display_name":"Vibhav Vineet","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vineet, Vibhav","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072006037","display_name":"Neel Joshi","orcid":"https://orcid.org/0000-0001-8236-3566","observed_orcids":["https://orcid.org/0000-0001-8236-3566"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Joshi, Neel","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5009166220","display_name":"Baharan Mirzasoleiman","orcid":"https://orcid.org/0000-0002-9374-9496","observed_orcids":["https://orcid.org/0000-0002-9374-9496"]},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mirzasoleiman, Baharan","raw_affiliation_strings":["UCLA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UCLA","institution_ids":["https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0,"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/vision-language-models","display_name":"vision-language models","score":0.9160000085830688},{"id":"https://openalex.org/keywords/spatial-reasoning","display_name":"spatial reasoning","score":0.8169999718666077},{"id":"https://openalex.org/keywords/llava-1-5","display_name":"LLaVA-1.5","score":0.7870000004768372},{"id":"https://openalex.org/keywords/data-filtering","display_name":"data filtering","score":0.4580000042915344},{"id":"https://openalex.org/keywords/fine-tuning","display_name":"fine-tuning","score":0.6230000257492065}],"concepts":[{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6304367780685425},{"id":"https://openalex.org/C91632574","wikidata":"https://www.wikidata.org/wiki/Q15088675","display_name":"Data curation","level":2,"score":0.5401874780654907},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4570581614971161},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3967043459415436},{"id":"https://openalex.org/C70721500","wikidata":"https://www.wikidata.org/wiki/Q177005","display_name":"Computational biology","level":1,"score":0.3315674662590027},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.264015793800354},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.2501835823059082},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.12576380372047424},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.07227754592895508}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2501.04155","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2501.04155","pdf_url":"https://arxiv.org/pdf/2501.04155","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2501.04155","doi":"https://doi.org/10.48550/arxiv.2501.04155","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2501.04155","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2501.04155","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2501.04155","pdf_url":"https://arxiv.org/pdf/2501.04155","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G6671297155","display_name":null,"funder_award_id":"CAREER","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7490364631","display_name":"CAREER: Coresets for Robust and Efficient Machine Learning","funder_award_id":"2146492","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8585541598","display_name":"NSF-Simons AI Institute for Cosmic Origins","funder_award_id":"2421782","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320307791","display_name":"Cisco Systems","ror":"https://ror.org/03yt1ez60"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4406231070.pdf","grobid_xml":"https://content.openalex.org/works/W4406231070.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4406191860","https://openalex.org/W4415539179","https://openalex.org/W4407819180","https://openalex.org/W4392781380","https://openalex.org/W4416016490","https://openalex.org/W4416424507","https://openalex.org/W4409760471","https://openalex.org/W7154034438","https://openalex.org/W6929447323","https://openalex.org/W4403908252"],"abstract_inverted_index":{"Vision-language":[0],"models":[1],"(VLMs)":[2],"are":[3],"highly":[4],"effective":[5],"but":[6],"often":[7],"underperform":[8],"on":[9,77,101,106],"specialized":[10,144],"tasks;":[11],"for":[12,41,56,109,124],"example,":[13],"Llava-1.5":[14,110],"struggles":[15],"with":[16,89],"chart":[17],"and":[18,80,84,104,136,143],"diagram":[19,107],"understanding":[20,108],"due":[21],"to":[22,35,95,113,120],"scarce":[23],"task-specific":[24,133],"training":[25,28],"data.":[26,86],"Existing":[27],"data,":[29,116],"sourced":[30],"from":[31],"general-purpose":[32,141],"datasets,":[33],"fails":[34],"capture":[36],"the":[37,125,138],"nuanced":[38],"details":[39],"needed":[40],"these":[42],"tasks.":[43],"We":[44],"introduce":[45],"MM-Gen,":[46],"a":[47,65],"scalable":[48],"method":[49],"that":[50],"generates":[51],"task-specific,":[52],"high-quality":[53],"synthetic":[54],"text":[55,75],"candidate":[57],"images":[58],"by":[59,92],"leveraging":[60],"stronger":[61],"models.":[62],"MM-Gen":[63,93,117],"employs":[64],"three-stage":[66],"targeted":[67,74],"process:":[68],"partitioning":[69],"data":[70,90],"into":[71],"subgroups,":[72],"generating":[73],"based":[76],"task":[78],"descriptions,":[79],"filtering":[81],"out":[82],"redundant":[83],"outlier":[85],"Fine-tuning":[87],"VLMs":[88],"generated":[91],"leads":[94],"significant":[96],"performance":[97,135],"gains,":[98],"including":[99],"29%":[100],"spatial":[102],"reasoning":[103],"15%":[105],"(7B).":[111],"Compared":[112],"human-curated":[114],"caption":[115],"achieves":[118],"up":[119],"1.6x":[121],"better":[122],"improvements":[123],"original":[126],"models,":[127],"proving":[128],"its":[129],"effectiveness":[130],"in":[131],"enhancing":[132],"VLM":[134],"bridging":[137],"gap":[139],"between":[140],"datasets":[142],"requirements.":[145],"Code":[146],"available":[147],"at":[148],"https://github.com/sjoshi804/MM-Gen.":[149]},"counts_by_year":[],"updated_date":"2026-10-17T09:40:24.721904","created_date":"2025-01-10T00:00:00"}