{"id":"https://openalex.org/W4412945126","doi":"https://doi.org/10.18653/v1/2025.acl-long.1198","title":"OMGM: Orchestrate Multiple Granularities and Modalities for Efficient Multimodal Retrieval","display_name":"OMGM: Orchestrate Multiple Granularities and Modalities for Efficient Multimodal Retrieval","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945126","doi":"https://doi.org/10.18653/v1/2025.acl-long.1198"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.1198","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1198","pdf_url":"https://aclanthology.org/2025.acl-long.1198.pdf","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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.1198.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100663925","display_name":"Wei Yang","orcid":"https://orcid.org/0000-0002-2979-3425"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Yang","raw_affiliation_strings":["Microsoft Research Asia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024800968","display_name":"Jingjing Fu","orcid":"https://orcid.org/0000-0003-1811-6637"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingjing Fu","raw_affiliation_strings":["Microsoft Research Asia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100706529","display_name":"Rui Wang","orcid":"https://orcid.org/0000-0001-9350-3667"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rui Wang","raw_affiliation_strings":["Microsoft Research Asia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100614565","display_name":"Jinyu Wang","orcid":"https://orcid.org/0000-0001-5593-9582"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinyu Wang","raw_affiliation_strings":["Microsoft Research Asia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059355028","display_name":"Lei Song","orcid":"https://orcid.org/0009-0004-8460-1926"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei Song","raw_affiliation_strings":["Microsoft Research Asia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5055104982","display_name":"\u5c06\u5c1a \u6e21\u8fba","orcid":"https://orcid.org/0000-0002-6337-9375"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiang Bian","raw_affiliation_strings":["Microsoft Research Asia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210113369"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"24545","last_page":"24563"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9850999712944031,"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"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9850999712944031,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9557999968528748,"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/computer-science","display_name":"Computer science","score":0.7732586860656738},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.719092845916748},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.5342724919319153},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.33737772703170776}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7732586860656738},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.719092845916748},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.5342724919319153},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33737772703170776},{"id":"https://openalex.org/C36289849","wikidata":"https://www.wikidata.org/wiki/Q34749","display_name":"Social science","level":1,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.1198","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1198","pdf_url":"https://aclanthology.org/2025.acl-long.1198.pdf","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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.1198","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1198","pdf_url":"https://aclanthology.org/2025.acl-long.1198.pdf","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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945126.pdf","grobid_xml":"https://content.openalex.org/works/W4412945126.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2185469136","https://openalex.org/W2011264131","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4306353150","https://openalex.org/W2026860389"],"abstract_inverted_index":{"Vision-language":[0,29],"retrieval-augmented":[1],"generation":[2],"(RAG)":[3],"has":[4],"become":[5],"an":[6],"effective":[7],"approach":[8],"for":[9,95,110,124],"tackling":[10],"Knowledge-Based":[11],"Visual":[12],"Question":[13],"Answering":[14],"(KB-VQA),":[15],"which":[16,36],"requires":[17],"external":[18],"knowledge":[19,46,52,93],"beyond":[20],"the":[21,42,60,106,119,129],"visual":[22],"content":[23],"presented":[24],"in":[25,48,149],"images.The":[26],"effectiveness":[27,148],"of":[28],"RAG":[30,69],"systems":[31],"hinges":[32],"on":[33,128],"multimodal":[34,68,101,108],"retrieval,":[35,97],"is":[37],"inherently":[38],"challenging":[39],"due":[40],"to":[41,82,104],"diverse":[43],"modalities":[44,81],"and":[45,51,80,131,141],"granularities":[47,79],"both":[49],"queries":[50],"bases.Existing":[53],"methods":[54],"have":[55],"not":[56],"fully":[57],"tapped":[58],"into":[59],"potential":[61],"interplay":[62],"between":[63],"these":[64],"elements.We":[65],"propose":[66],"a":[67,72,88,100],"system":[70,85],"featuring":[71],"coarse-to-fine,":[73],"multi-step":[74],"retrieval":[75,139],"that":[76],"harmonizes":[77],"multiple":[78],"enhance":[83],"efficacy.Our":[84],"begins":[86],"with":[87],"broad":[89],"initial":[90],"search":[91],"aligning":[92],"granularity":[94],"cross-modal":[96],"followed":[98],"by":[99],"fusion":[102],"reranking":[103],"capture":[105],"nuanced":[107],"information":[109],"top":[111],"entity":[112],"selection.A":[113],"text":[114],"reranker":[115],"then":[116],"filters":[117],"out":[118],"most":[120],"relevant":[121],"fine-grained":[122],"section":[123],"augmented":[125],"generation.Extensive":[126],"experiments":[127],"InfoSeek":[130],"Encyclopedic-VQA":[132],"benchmarks":[133],"show":[134],"our":[135],"method":[136],"achieves":[137],"state-of-the-art":[138],"performance":[140],"highly":[142],"competitive":[143],"answering":[144],"results,":[145],"underscoring":[146],"its":[147],"advancing":[150],"KB-VQA":[151],"systems.Our":[152],"code":[153],"can":[154],"be":[155],"found":[156],"at":[157],"https://github.com/ChaoLinAViy/OMGM.":[158]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}