{"id":"https://openalex.org/W4409047820","doi":"https://doi.org/10.1145/3721146.3721941","title":"Leveraging Approximate Caching for Faster Retrieval-Augmented Generation","display_name":"Leveraging Approximate Caching for Faster Retrieval-Augmented Generation","publication_year":2025,"publication_date":"2025-03-30","ids":{"openalex":"https://openalex.org/W4409047820","doi":"https://doi.org/10.1145/3721146.3721941"},"language":"en","primary_location":{"id":"doi:10.1145/3721146.3721941","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721146.3721941","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721941","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721941","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5034711959","display_name":"Shai Bergman","orcid":"https://orcid.org/0000-0002-3484-1452"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Shai Aviram Bergman","raw_affiliation_strings":["Huawei, Zurich, Switzerland"],"raw_orcid":"https://orcid.org/0000-0002-3484-1452","affiliations":[{"raw_affiliation_string":"Huawei, Zurich, Switzerland","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhang Ji","orcid":"https://orcid.org/0009-0008-6184-7279"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang Ji","raw_affiliation_strings":["Huawei, Zurich, Switzerland"],"raw_orcid":"https://orcid.org/0009-0008-6184-7279","affiliations":[{"raw_affiliation_string":"Huawei, Zurich, Switzerland","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031849486","display_name":"Anne-Marie Kermarrec","orcid":"https://orcid.org/0000-0001-8187-724X"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Anne-Marie Kermarrec","raw_affiliation_strings":["EPFL, Lausanne, Switzerland"],"raw_orcid":"https://orcid.org/0000-0001-8187-724X","affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043612282","display_name":"Diana Petrescu","orcid":null},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Diana Petrescu","raw_affiliation_strings":["EPFL, Lausanne, Switzerland"],"raw_orcid":"https://orcid.org/0009-0006-2229-235X","affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000635510","display_name":"Rafael Pires","orcid":"https://orcid.org/0000-0002-7826-1599"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Rafael Pires","raw_affiliation_strings":["EPFL, Lausanne, Switzerland"],"raw_orcid":"https://orcid.org/0000-0002-7826-1599","affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5116867120","display_name":"Mathis Randl","orcid":null},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Mathis Randl","raw_affiliation_strings":["EPFL, Lausanne, Switzerland"],"raw_orcid":"https://orcid.org/0009-0003-6844-4695","affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5010233454","display_name":"Martijn de Vos","orcid":"https://orcid.org/0000-0003-4157-4847"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Martijn de Vos","raw_affiliation_strings":["EPFL, Lausanne, Switzerland"],"raw_orcid":"https://orcid.org/0000-0003-4157-4847","affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5034711959"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":5.6812,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.9569001,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"66","last_page":"73"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11478","display_name":"Caching and Content Delivery","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11478","display_name":"Caching and Content Delivery","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9904000163078308,"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/T12288","display_name":"Optimization and Search Problems","score":0.9810000061988831,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.8127959370613098},{"id":"https://openalex.org/keywords/augmented-reality","display_name":"Augmented reality","score":0.4165477156639099},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.3972027003765106},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3158835172653198}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8127959370613098},{"id":"https://openalex.org/C153715457","wikidata":"https://www.wikidata.org/wiki/Q254183","display_name":"Augmented reality","level":2,"score":0.4165477156639099},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3972027003765106},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3158835172653198}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3721146.3721941","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721146.3721941","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721941","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2503.05530","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.05530","pdf_url":"https://arxiv.org/pdf/2503.05530","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3721146.3721941","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721146.3721941","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721941","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4409047820.pdf"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W2124509324","https://openalex.org/W2126356880","https://openalex.org/W2168467811","https://openalex.org/W2292999625","https://openalex.org/W2318810549","https://openalex.org/W2806803384","https://openalex.org/W2963469388","https://openalex.org/W3027879771","https://openalex.org/W4226082499","https://openalex.org/W4313331827","https://openalex.org/W4380322001","https://openalex.org/W4385474609","https://openalex.org/W4389519928","https://openalex.org/W4392019609","https://openalex.org/W4396822175","https://openalex.org/W4398757454","https://openalex.org/W4400641571","https://openalex.org/W4402827393","https://openalex.org/W4403899386","https://openalex.org/W4403977136","https://openalex.org/W4405942816","https://openalex.org/W6651372575","https://openalex.org/W6678775411"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2172197285","https://openalex.org/W2991048842","https://openalex.org/W2750280393","https://openalex.org/W2355696739","https://openalex.org/W3158001554","https://openalex.org/W2771909920","https://openalex.org/W2957704286"],"abstract_inverted_index":{"Retrieval-augmented":[0],"generation":[1],"(RAG)":[2],"enhances":[3],"the":[4,19,47,82,111,115,127],"reliability":[5],"of":[6,57],"large":[7,29],"language":[8],"model":[9],"(LLM)":[10],"answers":[11],"by":[12,50,102],"integrating":[13],"external":[14],"knowledge.":[15],"However,":[16],"RAG":[17,48],"increases":[18],"end-to-end":[20],"inference":[21],"time":[22],"since":[23],"looking":[24],"for":[25,145],"relevant":[26],"documents":[27,66],"from":[28],"vector":[30,75,116],"databases":[31],"is":[32,139],"computationally":[33],"expensive.":[34],"To":[35],"address":[36],"this,":[37],"we":[38],"introduce":[39],"Proximity,":[40],"an":[41],"approximate":[42,137],"key-value":[43],"cache":[44],"that":[45,88,136],"optimizes":[46],"workflow":[49],"leveraging":[51],"similarities":[52],"in":[53],"user":[54],"queries.":[55],"Instead":[56],"treating":[58],"each":[59],"query":[60],"independently,":[61],"Proximity":[62,80,98],"reuses":[63],"previously":[64],"retrieved":[65],"when":[67],"similar":[68],"queries":[69],"appear,":[70],"reducing":[71],"reliance":[72],"on":[73,81,114],"expensive":[74],"database":[76],"lookups.":[77],"We":[78,118],"evaluate":[79],"MMLU":[83],"and":[84,109,125,131,142],"MedRAG":[85],"benchmarks,":[86],"demonstrating":[87],"it":[89],"significantly":[90],"improves":[91],"retrieval":[92,100],"efficiency":[93],"while":[94,106],"maintaining":[95,107],"response":[96],"accuracy.":[97],"reduces":[99],"latency":[101],"up":[103],"to":[104],"59%":[105],"accuracy":[108],"lowers":[110],"computational":[112],"burden":[113],"database.":[117],"also":[119],"experiment":[120],"with":[121],"different":[122],"similarity":[123],"thresholds":[124],"quantify":[126],"trade-off":[128],"between":[129],"speed":[130],"recall.":[132],"Our":[133],"work":[134],"shows":[135],"caching":[138],"a":[140],"viable":[141],"effective":[143],"strategy":[144],"optimizing":[146],"RAG-based":[147],"systems.":[148]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":2}],"updated_date":"2026-06-05T09:01:59.212387","created_date":"2025-10-10T00:00:00"}
