{"id":"https://openalex.org/W7204148305","doi":"https://doi.org/10.48550/arxiv.2608.20845","title":"RAG Deserves an Index: Why Ingest-Time Compilation Beats Query-Time Interpretation","display_name":"RAG Deserves an Index: Why Ingest-Time Compilation Beats Query-Time Interpretation","publication_year":2026,"publication_date":"2026-08-21","ids":{"openalex":"https://openalex.org/W7204148305","doi":"https://doi.org/10.48550/arxiv.2608.20845"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2608.20845","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.20845","pdf_url":"https://arxiv.org/pdf/2608.20845","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.20845","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5147430606","display_name":"Kyle Wild","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I200641316","display_name":"Musashino University","ror":"https://ror.org/04bcbax71","country_code":"JP","type":"education","lineage":["https://openalex.org/I200641316"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Wild, Kyle","raw_affiliation_strings":["Endgame Labs, Inc. San Francisco , USA Asia AI Institute ,","Musashino University Tokyo , Japan orcid"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Endgame Labs, Inc. San Francisco , USA Asia AI Institute ,","institution_ids":[]},{"raw_affiliation_string":"Musashino University Tokyo , Japan orcid","institution_ids":["https://openalex.org/I200641316"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5147976931","display_name":"Yusuke Takahashi","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I200641316","display_name":"Musashino University","ror":"https://ror.org/04bcbax71","country_code":"JP","type":"education","lineage":["https://openalex.org/I200641316"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Takahashi, Yusuke","raw_affiliation_strings":["Faculty of Data Science , Musashino University Asia AI Institute AIx, Inc . Tokyo , Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Faculty of Data Science , Musashino University Asia AI Institute AIx, Inc . Tokyo , Japan","institution_ids":["https://openalex.org/I200641316"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5064335046","display_name":"Asako Uraki","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I200641316","display_name":"Musashino University","ror":"https://ror.org/04bcbax71","country_code":"JP","type":"education","lineage":["https://openalex.org/I200641316"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Uraki, Asako","raw_affiliation_strings":["Faculty of Data Science , Musashino University Asia AI Institute AIx, Inc . Tokyo , Japan orcid"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Faculty of Data Science , Musashino University Asia AI Institute AIx, Inc . Tokyo , Japan orcid","institution_ids":["https://openalex.org/I200641316"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I200641316"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.8036999702453613,"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":0.8036999702453613,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.17350000143051147,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.004399999976158142,"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/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.9520000219345093},{"id":"https://openalex.org/keywords/corpus-indexing","display_name":"corpus indexing","score":0.781000018119812},{"id":"https://openalex.org/keywords/inference-cost-reduction","display_name":"inference cost reduction","score":0.5429999828338623}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7353000044822693},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.527400016784668},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5199000239372253},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.48190000653266907},{"id":"https://openalex.org/C2780876879","wikidata":"https://www.wikidata.org/wiki/Q3054749","display_name":"Meaning (existential)","level":2,"score":0.4415999948978424},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4383000135421753},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4343000054359436},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.42480000853538513},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.41940000653266907},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4041000008583069},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4032999873161316},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.37290000915527344},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3580999970436096},{"id":"https://openalex.org/C108710211","wikidata":"https://www.wikidata.org/wiki/Q11538","display_name":"Mathematical proof","level":2,"score":0.3467000126838684},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3384000062942505},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.3197000026702881},{"id":"https://openalex.org/C554936623","wikidata":"https://www.wikidata.org/wiki/Q199657","display_name":"Reading (process)","level":2,"score":0.31679999828338623},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.3138999938964844},{"id":"https://openalex.org/C61423126","wikidata":"https://www.wikidata.org/wiki/Q187432","display_name":"Scripting language","level":2,"score":0.3091000020503998},{"id":"https://openalex.org/C98184364","wikidata":"https://www.wikidata.org/wiki/Q1780131","display_name":"Argument (complex analysis)","level":2,"score":0.29840001463890076},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.29580000042915344},{"id":"https://openalex.org/C93518851","wikidata":"https://www.wikidata.org/wiki/Q180160","display_name":"Metadata","level":2,"score":0.29499998688697815},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.29499998688697815},{"id":"https://openalex.org/C95318506","wikidata":"https://www.wikidata.org/wiki/Q6588467","display_name":"Textual entailment","level":3,"score":0.29109999537467957},{"id":"https://openalex.org/C2777526511","wikidata":"https://www.wikidata.org/wiki/Q691543","display_name":"Pace","level":2,"score":0.28859999775886536},{"id":"https://openalex.org/C5655090","wikidata":"https://www.wikidata.org/wiki/Q192588","display_name":"Relational database","level":2,"score":0.2833999991416931},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.2815000116825104},{"id":"https://openalex.org/C2778348673","wikidata":"https://www.wikidata.org/wiki/Q739302","display_name":"Production (economics)","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C200833197","wikidata":"https://www.wikidata.org/wiki/Q333707","display_name":"Compile time","level":3,"score":0.27720001339912415},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.2685999870300293},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C60048249","wikidata":"https://www.wikidata.org/wiki/Q37437","display_name":"Syntax","level":2,"score":0.26269999146461487},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C132964779","wikidata":"https://www.wikidata.org/wiki/Q2110223","display_name":"Raw data","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.2502000033855438}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.20845","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.20845","pdf_url":"https://arxiv.org/pdf/2608.20845","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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"},{"id":"doi:10.48550/arxiv.2608.20845","doi":"https://doi.org/10.48550/arxiv.2608.20845","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.20845","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.20845","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.20845","pdf_url":"https://arxiv.org/pdf/2608.20845","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.750272810459137}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204148305.pdf","grobid_xml":"https://content.openalex.org/works/W7204148305.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Nearly":[0],"every":[1],"retrieval-augmented":[2],"question-answering":[3],"system":[4],"in":[5],"production":[6],"ships":[7],"with":[8,130,158,177,243,278],"a":[9,15,88,106,123,127,154,198,240],"hidden":[10],"interpreter:":[11],"on":[12,197],"each":[13],"query":[14],"language":[16],"model":[17],"re-derives":[18],"the":[19,36,61,65,69,72,80,116,208,228,257,279],"meaning":[20,125],"of":[21,44,64,201],"raw":[22],"corpus":[23,96,181],"text":[24],"and":[25,68,109,138,149,166,246,261],"then":[26],"throws":[27],"that":[28,91,151,236,264],"work":[29,82],"away.":[30],"Cheaper":[31],"models":[32],"do":[33,79],"not":[34],"close":[35,277],"gap:":[37],"per-token":[38],"prices":[39,57],"have":[40],"fallen":[41],"by":[42],"orders":[43],"magnitude":[45],"while":[46,190],"inference":[47],"spend":[48],"has":[49,271],"risen,":[50],"because":[51,269],"context":[52],"volume":[53],"grows":[54],"faster":[55],"than":[56,180,188],"fall.":[58],"This":[59],"is":[60,71,100,143,239],"modern":[62],"equivalent":[63],"full-table":[66],"scan,":[67],"remedy":[70],"one":[73],"databases":[74],"found":[75],"fifty":[76],"years":[77],"ago:":[78],"expensive":[81],"once,":[83],"at":[84,145,253],"write":[85],"time,":[86],"into":[87,126],"maintained":[89,136],"structure":[90],"makes":[92],"reads":[93],"cheap.":[94],"A":[95],"whose":[97,141],"read":[98,288],"pattern":[99],"known":[101],"before":[102],"it":[103,192,262,270],"ever":[104],"meets":[105],"user":[107],"can":[108],"should":[110],"be":[111],"indexed":[112],"too.":[113],"We":[114,276],"call":[115],"paradigm":[117],"ingest-time":[118],"semantic":[119],"compilation":[120,285],"(ISC):":[121],"compile":[122,146],"corpus's":[124],"queryable":[128],"substrate":[129,152],"two":[131],"coupled":[132],"layers":[133],"-":[134,148,260],"incrementally":[135],"embeddings,":[137],"atomic":[139],"claims":[140,206,252],"provenance":[142],"validated":[144],"time":[147],"treat":[150],"as":[153,207],"first-class":[155],"database":[156],"object":[157],"its":[159],"own":[160],"DDL,":[161],"maintenance":[162],"contract,":[163,165],"migration":[164],"cost":[167],"model.":[168],"Two":[169],"existence":[170],"proofs":[171],"support":[172],"it.":[173],"Substrate":[174],"upkeep":[175],"scales":[176],"change":[178],"rather":[179],"size:":[182],"incremental":[183],"updates":[184],"run":[185],"33.7x":[186],"cheaper":[187],"reconstruction":[189],"tracking":[191],"to":[193,274,287],"floating-point":[194],"precision.":[195],"And":[196],"held-out":[199],"sample":[200],"500":[202],"broadcast-interview":[203],"transcripts,":[204],"compiled":[205,251],"retrieval":[209,245],"payload":[210],"win":[211],"all":[212],"32":[213],"budget-by-model":[214],"cells:":[215],"85.2%":[216],"correct":[217],"from":[218,225,250,284],"roughly":[219,254],"2.2k":[220],"reader":[221],"tokens":[222,259],"against":[223],"72.5%":[224],"16.3k":[226],"for":[227],"best":[229],"chunk":[230],"configuration":[231],"anywhere.":[232],"The":[233],"only":[234],"baseline":[235],"keeps":[237],"pace":[238],"contextualized-chunk":[241],"pipeline":[242],"hybrid":[244],"reranking,":[247],"statistically":[248],"indistinguishable":[249],"twenty-one":[255],"times":[256],"query-path":[258],"reaches":[263],"parity,":[265],"we":[266],"argue,":[267],"precisely":[268],"itself":[272],"begun":[273],"compile.":[275],"systems":[280],"agenda":[281],"this":[282],"opens,":[283],"planners":[286],"planning.":[289]},"counts_by_year":[],"updated_date":"2026-10-12T09:40:04.860316","created_date":"2026-08-25T00:00:00"}