{"id":"https://openalex.org/W7163521779","doi":"https://doi.org/10.23919/date69613.2026.11539352","title":"FaTRQ: Tiered Residual Quantization for LLM Vector Search in Far-Memory-Aware ANNS Systems","display_name":"FaTRQ: Tiered Residual Quantization for LLM Vector Search in Far-Memory-Aware ANNS Systems","publication_year":2026,"publication_date":"2026-04-20","ids":{"openalex":"https://openalex.org/W7163521779","doi":"https://doi.org/10.23919/date69613.2026.11539352"},"language":null,"primary_location":{"id":"doi:10.23919/date69613.2026.11539352","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date69613.2026.11539352","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 Design, Automation & Test in Europe Conference (DATE)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5084707472","display_name":"T Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tianqi Zhang","raw_affiliation_strings":["UC San Diego,La Jolla,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC San Diego,La Jolla,CA,USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082759711","display_name":"Flavio Ponzina","orcid":"https://orcid.org/0000-0002-9662-498X"},"institutions":[{"id":"https://openalex.org/I26538001","display_name":"San Diego State University","ror":"https://ror.org/0264fdx42","country_code":"US","type":"education","lineage":["https://openalex.org/I26538001"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Flavio Ponzina","raw_affiliation_strings":["San Diego State University,San Diego,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"San Diego State University,San Diego,CA,USA","institution_ids":["https://openalex.org/I26538001"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5137819923","display_name":"Tajana Rosing","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tajana Rosing","raw_affiliation_strings":["UC San Diego,La Jolla,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC San Diego,La Jolla,CA,USA","institution_ids":["https://openalex.org/I36258959"]}]}],"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":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.11890000104904175,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.11890000104904175,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.11559999734163284,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.09880000352859497,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.5098999738693237},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4754999876022339},{"id":"https://openalex.org/keywords/vector-quantization","display_name":"Vector quantization","score":0.428600013256073},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.32350000739097595},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.31310001015663147},{"id":"https://openalex.org/keywords/control-theory","display_name":"Control theory (sociology)","score":0.298799991607666}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5722000002861023},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.5098999738693237},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5063999891281128},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4754999876022339},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.428600013256073},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4235999882221222},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.32350000739097595},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.31310001015663147},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.31049999594688416},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.298799991607666},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.2962000072002411},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.26759999990463257},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2671000063419342},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25760000944137573},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2531000077724457}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.23919/date69613.2026.11539352","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date69613.2026.11539352","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 Design, Automation & Test in Europe Conference (DATE)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W2034861439","https://openalex.org/W2041811089","https://openalex.org/W2077815765","https://openalex.org/W2124509324","https://openalex.org/W2133995768","https://openalex.org/W2346205343","https://openalex.org/W2963469388","https://openalex.org/W3042631625","https://openalex.org/W3094858795","https://openalex.org/W3174809957","https://openalex.org/W4386763970","https://openalex.org/W4387064022","https://openalex.org/W4399175194","https://openalex.org/W4400909749","https://openalex.org/W4407361744","https://openalex.org/W4415124086","https://openalex.org/W4416203810","https://openalex.org/W7133250524","https://openalex.org/W7134928809"],"related_works":[],"abstract_inverted_index":{"Approximate":[0],"Nearest-Neighbor":[1],"Search":[2],"(ANNS)":[3],"is":[4,120,147],"a":[5,47,80,99,118,150],"key":[6],"technique":[7],"in":[8,39,141,149],"retrieval-augmented":[9],"generation":[10],"(RAG),":[11],"enabling":[12],"rapid":[13],"identification":[14],"of":[15,73],"the":[16,71,74,89,123,162,169],"most":[17],"relevant":[18],"high-dimensional":[19],"embeddings":[20],"from":[21,56,95,111],"massive":[22],"vector":[23],"databases.":[24],"Modern":[25],"ANNS":[26,178],"engines":[27],"accelerate":[28],"this":[29],"process":[30],"using":[31,84,107],"prebuilt":[32],"indexes":[33],"and":[34,64,167],"store":[35],"compressed":[36],"vector-quantized":[37],"representations":[38],"fast":[40],"memory.":[41,113,143],"However,":[42],"they":[43],"still":[44],"rely":[45],"on":[46],"costly":[48],"second-pass":[49],"refinement":[50,82,157],"stage":[51],"that":[52,87,103],"reads":[53,68],"full-precision":[54],"vectors":[55,94],"slower":[57],"storage":[58,163],"like":[59],"SSDs.":[60],"For":[61],"modern":[62],"text":[63],"multimodal":[65],"embeddings,":[66],"these":[67],"now":[69],"dominate":[70],"latency":[72],"entire":[75],"query.":[76],"We":[77],"propose":[78,129],"FaTRQ,":[79],"far-memory-aware":[81],"system":[83],"tiered":[85,130],"memory":[86],"eliminates":[88],"need":[90],"to":[91,154,173],"fetch":[92],"full":[93],"storage.":[96],"It":[97],"introduces":[98],"progressive":[100],"distance":[101],"estimator":[102],"refines":[104],"coarse":[105],"scores":[106],"compact":[108],"residuals":[109,135],"streamed":[110],"far":[112,142],"Refinement":[114],"stops":[115],"early":[116],"once":[117],"candidate":[119],"provably":[121],"outside":[122],"top-k.":[124],"To":[125],"support":[126],"this,":[127],"we":[128],"residual":[131],"quantization,":[132],"which":[133],"encodes":[134],"as":[136],"ternary":[137],"values":[138],"stored":[139],"efficiently":[140],"A":[144],"custom":[145],"accelerator":[146],"deployed":[148],"CXL":[151],"Type-2":[152],"device":[153],"perform":[155],"low-latency":[156],"locally.":[158],"Together,":[159],"FaTRQ":[160],"improves":[161,168],"efficiency":[164],"by":[165,171],"2.4\u00d7":[166],"throughput":[170],"up":[172],"9\u00d7":[174],"than":[175],"SOTA":[176],"GPU":[177],"system.":[179]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-06-05T00:00:00"}