{"id":"https://openalex.org/W7108315616","doi":"https://doi.org/10.1145/3767695.3769486","title":"CogPlanner: Unveiling the Potential of Agentic Multimodal Retrieval Augmented Generation with Planning","display_name":"CogPlanner: Unveiling the Potential of Agentic Multimodal Retrieval Augmented Generation with Planning","publication_year":2025,"publication_date":"2025-12-03","ids":{"openalex":"https://openalex.org/W7108315616","doi":"https://doi.org/10.1145/3767695.3769486"},"language":null,"primary_location":{"id":"doi:10.1145/3767695.3769486","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767695.3769486","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 2025 Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3767695.3769486","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Xiaohan Yu","orcid":"https://orcid.org/0009-0005-3560-1230"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Xiaohan Yu","raw_affiliation_strings":["Huawei Cloud BU, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0005-3560-1230","affiliations":[{"raw_affiliation_string":"Huawei Cloud BU, Beijing, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhihan Yang","orcid":"https://orcid.org/0009-0002-0600-5833"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihan Yang","raw_affiliation_strings":["Huawei Cloud BU, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0002-0600-5833","affiliations":[{"raw_affiliation_string":"Huawei Cloud BU, Beijing, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"last","author":{"id":null,"display_name":"Chong Chen","orcid":"https://orcid.org/0000-0003-1417-2295"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chong Chen","raw_affiliation_strings":["Huawei Cloud BU, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1417-2295","affiliations":[{"raw_affiliation_string":"Huawei Cloud BU, Beijing, China","institution_ids":["https://openalex.org/I2250955327"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I2250955327"],"apc_list":null,"apc_paid":null,"fwci":1.1332,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.85202528,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"2","last_page":"11"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.698199987411499,"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":0.698199987411499,"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/T10028","display_name":"Topic Modeling","score":0.11110000312328339,"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/T12031","display_name":"Speech and dialogue systems","score":0.08789999783039093,"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/task","display_name":"Task (project management)","score":0.6959999799728394},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.6542999744415283},{"id":"https://openalex.org/keywords/cognitive-models-of-information-retrieval","display_name":"Cognitive models of information retrieval","score":0.4433000087738037},{"id":"https://openalex.org/keywords/multimodality","display_name":"Multimodality","score":0.3513000011444092},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.33230000734329224},{"id":"https://openalex.org/keywords/information-integration","display_name":"Information integration","score":0.3301999866962433}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8230000138282776},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6959999799728394},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6542999744415283},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5192999839782715},{"id":"https://openalex.org/C21025794","wikidata":"https://www.wikidata.org/wiki/Q5141219","display_name":"Cognitive models of information retrieval","level":4,"score":0.4433000087738037},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3928000032901764},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.38190001249313354},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.3513000011444092},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.33230000734329224},{"id":"https://openalex.org/C33326189","wikidata":"https://www.wikidata.org/wiki/Q17092450","display_name":"Information integration","level":2,"score":0.3301999866962433},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.3294000029563904},{"id":"https://openalex.org/C99016210","wikidata":"https://www.wikidata.org/wiki/Q5488129","display_name":"Query expansion","level":2,"score":0.32829999923706055},{"id":"https://openalex.org/C135641252","wikidata":"https://www.wikidata.org/wiki/Q738567","display_name":"Multimodal interaction","level":2,"score":0.3009999990463257},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.28929999470710754},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.26579999923706055}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3767695.3769486","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767695.3769486","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 2025 Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3767695.3769486","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767695.3769486","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 2025 Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth","score":0.5322787761688232}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W2964138343","https://openalex.org/W4387969743","https://openalex.org/W4402667093","https://openalex.org/W4402916777"],"related_works":[],"abstract_inverted_index":{"Multimodal":[0,51],"Retrieval":[1,52],"Augmented":[2,53],"Generation":[3,54],"(MRAG)":[4],"systems":[5],"have":[6],"shown":[7],"promise":[8],"in":[9,143],"enhancing":[10],"the":[11,48,116],"generation":[12],"capabilities":[13],"of":[14,37,50],"multimodal":[15],"large":[16],"language":[17],"models":[18],"(MLLMs).":[19],"However,":[20],"existing":[21,138],"MRAG":[22,117,139],"frameworks":[23],"primarily":[24],"adhere":[25],"to":[26,33,91,113],"rigid,":[27],"single-step":[28],"retrieval":[29,89],"strategies":[30,90],"that":[31,58,134],"fail":[32],"address":[34],"real-world":[35],"challenges":[36],"information":[38,62],"acquisition":[39],"and":[40,64,88,94,101,120,146],"query":[41,86],"reformulation.":[42],"In":[43],"this":[44],"work,":[45],"we":[46,71,106],"introduce":[47,107],"task":[49,119],"Planning":[55,118],"(MRAG":[56],"Planning)":[57],"aims":[59],"at":[60],"effective":[61],"seeking":[63],"integration":[65,124],"while":[66],"minimizing":[67],"computational":[68,151],"overhead.":[69],"Specifically,":[70],"propose":[72],"CogPlanner,":[73],"an":[74],"agentic":[75],"plug-and-play":[76],"framework":[77],"inspired":[78],"by":[79],"human":[80],"cognitive":[81],"processes,":[82],"which":[83],"iteratively":[84],"determines":[85],"reformulation":[87],"generate":[92],"accurate":[93],"contextually":[95],"relevant":[96],"responses.":[97],"CogPlanner":[98,123,135],"supports":[99],"parallel":[100],"sequential":[102],"modeling":[103],"paradigms.":[104],"Furthermore,":[105],"CogBench,":[108],"a":[109],"new":[110],"benchmark":[111],"designed":[112],"rigorously":[114],"evaluate":[115],"facilitate":[121],"lightweight":[122],"with":[125,148],"resource-efficient":[126],"MLLMs,":[127],"such":[128],"as":[129],"Qwen2-VL-7B-Cog.":[130],"Experimental":[131],"results":[132],"demonstrate":[133],"significantly":[136],"outperforms":[137],"baselines,":[140],"offering":[141],"improvements":[142],"both":[144],"accuracy":[145],"efficiency":[147],"minimal":[149],"additional":[150],"costs.":[152]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-12-04T23:47:47.292601","created_date":"2025-12-03T00:00:00"}