{"id":"https://openalex.org/W7155377833","doi":"https://doi.org/10.1109/tcsvt.2026.3686876","title":"FineG-RAG: Fine-Grained Retrieval-Augmented Generation for Multimodal Large Language Models","display_name":"FineG-RAG: Fine-Grained Retrieval-Augmented Generation for Multimodal Large Language Models","publication_year":2026,"publication_date":"2026-04-23","ids":{"openalex":"https://openalex.org/W7155377833","doi":"https://doi.org/10.1109/tcsvt.2026.3686876"},"language":null,"primary_location":{"id":"doi:10.1109/tcsvt.2026.3686876","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsvt.2026.3686876","pdf_url":null,"source":{"id":"https://openalex.org/S115173108","display_name":"IEEE Transactions on Circuits and Systems for Video Technology","issn_l":"1051-8215","issn":["1051-8215","1558-2205"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Video Technology","raw_type":"journal-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/A5101923561","display_name":"Lu Jin","orcid":"https://orcid.org/0000-0002-2964-426X"},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lu Jin","raw_affiliation_strings":["School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-2964-426X","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China","institution_ids":["https://openalex.org/I36399199"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134367277","display_name":"Cheng Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Lin","raw_affiliation_strings":["School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0002-6682-0338","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China","institution_ids":["https://openalex.org/I36399199"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077029839","display_name":"Xinguang Xiang","orcid":"https://orcid.org/0000-0002-2344-6174"},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinguang Xiang","raw_affiliation_strings":["School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-2344-6174","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China","institution_ids":["https://openalex.org/I36399199"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101812029","display_name":"Yanpeng Sun","orcid":"https://orcid.org/0000-0001-6249-5596"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yanpeng Sun","raw_affiliation_strings":["NExT++ Laboratory, School of Computing, National University of Singapore, Queenstown, Singapore"],"raw_orcid":"https://orcid.org/0000-0001-6249-5596","affiliations":[{"raw_affiliation_string":"NExT++ Laboratory, School of Computing, National University of Singapore, Queenstown, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017096005","display_name":"Zechao Li","orcid":"https://orcid.org/0000-0002-5341-5985"},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zechao Li","raw_affiliation_strings":["School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-5341-5985","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China","institution_ids":["https://openalex.org/I36399199"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035112538","display_name":"Jinhui Tang","orcid":"https://orcid.org/0000-0001-9008-222X"},"institutions":[{"id":"https://openalex.org/I167027274","display_name":"Nanjing Forestry University","ror":"https://ror.org/03m96p165","country_code":"CN","type":"education","lineage":["https://openalex.org/I167027274"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinhui Tang","raw_affiliation_strings":["Nanjing Forestry University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0001-9008-222X","affiliations":[{"raw_affiliation_string":"Nanjing Forestry University, Nanjing, China","institution_ids":["https://openalex.org/I167027274"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":2645,"currency":"USD","value_usd":2645},"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.3912307,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"36","issue":"6","first_page":"8675","last_page":"8686"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.47269999980926514,"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/T10028","display_name":"Topic Modeling","score":0.47269999980926514,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.11779999732971191,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.06729999929666519,"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/natural-language","display_name":"Natural language","score":0.36890000104904175},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.3402999937534332},{"id":"https://openalex.org/keywords/context-model","display_name":"Context model","score":0.3133000135421753},{"id":"https://openalex.org/keywords/language-identification","display_name":"Language identification","score":0.30959999561309814},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.28450000286102295},{"id":"https://openalex.org/keywords/natural-language-generation","display_name":"Natural language generation","score":0.27399998903274536}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7178000211715698},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45980000495910645},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.44290000200271606},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.36890000104904175},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.3402999937534332},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.3133000135421753},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.30959999561309814},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.27399998903274536},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.27070000767707825},{"id":"https://openalex.org/C3020028006","wikidata":"https://www.wikidata.org/wiki/Q9158","display_name":"Electronic mail","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.25099998712539673},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcsvt.2026.3686876","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsvt.2026.3686876","pdf_url":null,"source":{"id":"https://openalex.org/S115173108","display_name":"IEEE Transactions on Circuits and Systems for Video Technology","issn_l":"1051-8215","issn":["1051-8215","1558-2205"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Video Technology","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1186749179","display_name":null,"funder_award_id":"62372233","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2531279109","display_name":null,"funder_award_id":"62425603","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6750246608","display_name":null,"funder_award_id":"BK20240011","funder_id":"https://openalex.org/F4320334982","funder_display_name":"Basic Research Program of Jiangsu Province"},{"id":"https://openalex.org/G675028838","display_name":null,"funder_award_id":"BG2024042","funder_id":"https://openalex.org/F4320336616","funder_display_name":"Science and Technology Program of Gansu Province"},{"id":"https://openalex.org/G8263474416","display_name":null,"funder_award_id":"62272230","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation 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/F4320334982","display_name":"Basic Research Program of Jiangsu Province","ror":null},{"id":"https://openalex.org/F4320336616","display_name":"Science and Technology Program of Gansu Province","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W2138011018","https://openalex.org/W2533598788","https://openalex.org/W2737725206","https://openalex.org/W2763070548","https://openalex.org/W2954924421","https://openalex.org/W3139434170","https://openalex.org/W3176278386","https://openalex.org/W4313009108","https://openalex.org/W4321021073","https://openalex.org/W4382998913","https://openalex.org/W4386065803","https://openalex.org/W4386076004","https://openalex.org/W4387682352","https://openalex.org/W4388936587","https://openalex.org/W4389523839","https://openalex.org/W4391249049","https://openalex.org/W4393154594","https://openalex.org/W4399110646","https://openalex.org/W4402713111","https://openalex.org/W4402716288","https://openalex.org/W4402754134","https://openalex.org/W4402770330","https://openalex.org/W4402774103","https://openalex.org/W4402916777","https://openalex.org/W4403081466","https://openalex.org/W4404238284","https://openalex.org/W4404781941","https://openalex.org/W4405755238","https://openalex.org/W4406890407","https://openalex.org/W4413120602","https://openalex.org/W7133196460","https://openalex.org/W7133220561","https://openalex.org/W7133224126","https://openalex.org/W7133231369","https://openalex.org/W7133231679"],"related_works":[],"abstract_inverted_index":{"Fine-grained":[0],"visual":[1,109],"recognition":[2,78,89,150,177],"refers":[3],"to":[4,7,52,54,74,138],"the":[5,76,88,148],"ability":[6],"distinguish":[8],"subtle":[9,55],"differences":[10],"between":[11],"visually":[12],"similar":[13],"objects\u2014":[14],"a":[15,69,92,129,156,180],"fundamental":[16],"yet":[17],"challenging":[18],"capability":[19],"for":[20,59],"Multimodal":[21],"Large":[22],"Language":[23],"Models":[24],"(MLLMs).":[25],"In":[26],"this":[27,64,124],"paper,":[28],"we":[29,66,98,154],"observe":[30],"that":[31,169],"even":[32],"strong":[33],"open-source":[34,184],"MLLMs,":[35,153],"such":[36],"as":[37],"Qwen2-VL":[38],"and":[39,111,126],"InternVL2,":[40],"still":[41],"struggle":[42],"with":[43,107],"accurately":[44],"identifying":[45],"fine-grained":[46,77,85,100,119,149,164],"categories.":[47],"These":[48],"models":[49],"often":[50],"fail":[51],"attend":[53],"but":[56],"critical":[57],"details":[58],"precise":[60],"discrimination.":[61],"To":[62,95,145],"unlock":[63],"potential,":[65],"propose":[67],"FineG-RAG,":[68],"retrieval-augmented":[70],"generation":[71],"pipeline":[72],"designed":[73],"enhance":[75],"capabilities":[79,151],"of":[80,152,182],"MLLMs.":[81,185],"FineG-RAG":[82,170],"integrates":[83],"external":[84],"knowledge":[86,102,120],"into":[87,128],"process":[90],"via":[91],"generalized":[93],"retriever.":[94],"support":[96],"this,":[97],"construct":[99],"visual-language":[101,130],"database":[103,125],"containing":[104],"representative":[105],"images":[106],"wide":[108],"diversity":[110],"expert-crafted":[112],"attribute":[113],"descriptions":[114],"from":[115,123],"multiple":[116],"perspectives.":[117],"Relevant":[118],"is":[121],"retrieved":[122],"fed":[127],"augmented":[131],"prompt,":[132],"which":[133],"provides":[134],"rich":[135],"multimodal":[136],"context":[137],"guide":[139],"MLLMs":[140],"in":[141],"generating":[142],"accurate":[143],"labels.":[144],"better":[146],"evaluate":[147],"design":[155],"multiple-choice":[157],"evaluation":[158],"strategy":[159],"based":[160],"on":[161],"publicly":[162],"four":[163],"datasets.":[165],"Extensive":[166],"experiments":[167],"demonstrate":[168],"consistently":[171],"outperforms":[172],"baseline":[173],"methods,":[174],"achieving":[175],"superior":[176],"accuracy":[178],"across":[179],"range":[181],"off-the-shelf,":[183]},"counts_by_year":[],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2026-04-24T00:00:00"}