{"id":"https://openalex.org/W7168422056","doi":"https://doi.org/10.1145/3805712.3809870","title":"Latent Retrieval Augmented Generation","display_name":"Latent Retrieval Augmented Generation","publication_year":2026,"publication_date":"2026-07-15","ids":{"openalex":"https://openalex.org/W7168422056","doi":"https://doi.org/10.1145/3805712.3809870"},"language":null,"primary_location":{"id":"doi:10.1145/3805712.3809870","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805712.3809870","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805712.3809870","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140790666","display_name":"Shu Zhou","orcid":"https://orcid.org/0000-0001-5935-3159"},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shu Zhou","raw_affiliation_strings":["School of Information Management, Nanjing University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0001-5935-3159","affiliations":[{"raw_affiliation_string":"School of Information Management, Nanjing University, Nanjing, China","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140813276","display_name":"Junan Chen","orcid":"https://orcid.org/0009-0009-4646-4875"},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junan Chen","raw_affiliation_strings":["School of Information Management, Nanjing University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0009-4646-4875","affiliations":[{"raw_affiliation_string":"School of Information Management, Nanjing University, Nanjing, China","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140817524","display_name":"Rui Ling","orcid":"https://orcid.org/0009-0000-6734-560X"},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rui Ling","raw_affiliation_strings":["School of Information Management, Nanjing University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0000-6734-560X","affiliations":[{"raw_affiliation_string":"School of Information Management, Nanjing University, Nanjing, China","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140755823","display_name":"Tao Fan","orcid":"https://orcid.org/0000-0002-6846-2901"},"institutions":[{"id":"https://openalex.org/I137056471","display_name":"Nanjing University of Finance and Economics","ror":"https://ror.org/031y8am81","country_code":"CN","type":"education","lineage":["https://openalex.org/I137056471"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tao Fan","raw_affiliation_strings":["Nanjing University of Finance and Economics, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-6846-2901","affiliations":[{"raw_affiliation_string":"Nanjing University of Finance and Economics, Nanjing, China","institution_ids":["https://openalex.org/I137056471"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5140832328","display_name":"Hao Wang","orcid":"https://orcid.org/0000-0002-0131-0823"},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Wang","raw_affiliation_strings":["School of Information Management, Nanjing University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-0131-0823","affiliations":[{"raw_affiliation_string":"School of Information Management, Nanjing University, Nanjing, China","institution_ids":["https://openalex.org/I881766915"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4373","last_page":"4378"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":null,"topics":[],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5843999981880188},{"id":"https://openalex.org/keywords/parametric-statistics","display_name":"Parametric statistics","score":0.5490999817848206},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5331000089645386},{"id":"https://openalex.org/keywords/encode","display_name":"ENCODE","score":0.48829999566078186},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.47360000014305115},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.4433000087738037},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.3896999955177307},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.3578999936580658},{"id":"https://openalex.org/keywords/dual","display_name":"Dual (grammatical number)","score":0.32850000262260437}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7993999719619751},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5843999981880188},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.5490999817848206},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5331000089645386},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.48829999566078186},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.47360000014305115},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4659000039100647},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4433000087738037},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40230000019073486},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.3896999955177307},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3578999936580658},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.32850000262260437},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.32519999146461487},{"id":"https://openalex.org/C170133592","wikidata":"https://www.wikidata.org/wiki/Q1806883","display_name":"Latent semantic analysis","level":2,"score":0.3197999894618988},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.31279999017715454},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.304500013589859},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.29100000858306885},{"id":"https://openalex.org/C24574437","wikidata":"https://www.wikidata.org/wiki/Q7135228","display_name":"Parametric model","level":3,"score":0.28999999165534973},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2809000015258789},{"id":"https://openalex.org/C159423971","wikidata":"https://www.wikidata.org/wiki/Q177251","display_name":"Associative property","level":2,"score":0.27320000529289246},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.26989999413490295},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.26159998774528503},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.26089999079704285},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.25699999928474426},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3805712.3809870","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805712.3809870","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3805712.3809870","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805712.3809870","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":9,"referenced_works":["https://openalex.org/W2964120615","https://openalex.org/W4385570777","https://openalex.org/W4389519118","https://openalex.org/W4391876619","https://openalex.org/W4402671152","https://openalex.org/W4410466068","https://openalex.org/W4412377217","https://openalex.org/W4415795887","https://openalex.org/W7106263165"],"related_works":[],"abstract_inverted_index":{"Retrieval-augmented":[0],"generation":[1],"(RAG)":[2],"has":[3],"emerged":[4],"as":[5],"a":[6,119,155],"promising":[7],"solution":[8],"to":[9,188,219],"enhance":[10],"the":[11,37,128,150,164],"reliability":[12],"of":[13],"large":[14],"language":[15],"models":[16],"(LLMs)":[17],"with":[18,64,149,213],"external":[19],"knowledge.":[20],"Existing":[21],"RAG":[22,44,176,195,200],"methods":[23,30,41,218],"operate":[24],"in":[25,36,70,163],"explicit":[26,95,170],"representation":[27],"spaces:":[28],"in-context":[29,215],"inject":[31],"knowledge":[32,109,124,180],"through":[33,140],"text":[34],"tokens":[35,98],"input,":[38],"while":[39,76,182],"parametric":[40,217],"like":[42],"Parametric":[43,74,189],"encode":[45],"documents":[46,135],"into":[47,136],"model":[48],"parameters.":[49],"Although":[50],"effective,":[51],"these":[52,113],"approaches":[53],"face":[54],"inherent":[55],"limitations.":[56],"In-context":[57],"injection":[58,125,157],"suffers":[59],"from":[60],"quadratic":[61],"computational":[62],"complexity":[63],"context":[65],"length":[66],"and":[67,84,106,145,205,216],"degraded":[68],"performance":[69],"complex":[71],"reasoning":[72],"tasks.":[73],"injection,":[75],"reducing":[77],"inference":[78],"costs,":[79],"requires":[80],"substantial":[81],"storage":[82,185],"overhead":[83],"computationally":[85],"expensive":[86],"offline":[87,142],"preprocessing.":[88],"More":[89],"fundamentally,":[90],"both":[91,203],"paradigms":[92],"rely":[93],"on":[94,193],"discrete":[96],"representations":[97,139],"or":[99,172],"parameters":[100],"that":[101,122,198],"may":[102],"introduce":[103,116],"information":[104],"bottlenecks":[105],"hinder":[107],"seamless":[108],"integration.":[110],"To":[111],"address":[112],"challenges,":[114],"we":[115],"Latent":[117,175,199],"RAG,":[118],"novel":[120],"paradigm":[121],"performs":[123],"entirely":[126],"within":[127],"continuous":[129],"latent":[130,138,166],"space.":[131],"Our":[132],"approach":[133],"encodes":[134],"ultra-compact":[137],"an":[141],"compression":[143],"phase,":[144],"directly":[146],"fuses":[147],"them":[148],"LLM's":[151],"hidden":[152],"states":[153],"via":[154],"learned":[156],"mechanism":[158],"during":[159],"inference.":[160],"By":[161],"operating":[162],"semantic":[165],"space":[167],"rather":[168],"than":[169],"token":[171],"parameter":[173],"spaces,":[174],"enables":[177],"more":[178],"natural":[179],"integration":[181],"achieving":[183],"9,200X":[184],"reduction":[186],"compared":[187],"RAG.":[190],"Experimental":[191],"results":[192],"multiple":[194],"benchmarks":[196],"demonstrate":[197],"substantially":[201],"enhances":[202],"effectiveness":[204],"efficiency.":[206],"Furthermore,":[207],"it":[208],"can":[209],"be":[210],"seamlessly":[211],"combined":[212],"existing":[214],"achieve":[220],"even":[221],"better":[222],"performance.":[223]},"counts_by_year":[],"updated_date":"2026-07-17T05:52:16.776730","created_date":"2026-07-16T00:00:00"}