{"id":"https://openalex.org/W7164047557","doi":"https://doi.org/10.48550/arxiv.2606.07923","title":"Larch: Learned Query Optimization for Semantic Predicates","display_name":"Larch: Learned Query Optimization for Semantic Predicates","publication_year":2026,"publication_date":"2026-06-06","ids":{"openalex":"https://openalex.org/W7164047557","doi":"https://doi.org/10.48550/arxiv.2606.07923"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.07923","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.07923","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.07923","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5020064057","display_name":"Fuheng Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Fuheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138202636","display_name":"Pawel Liskowski","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liskowski, Pawel","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138281320","display_name":"Zihan Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Zihan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135466293","display_name":"Benjamin Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Benjamin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090488104","display_name":"Puxuan Yu","orcid":"https://orcid.org/0000-0001-7913-8632"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Puxuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138214950","display_name":"Varich Boonsanong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Boonsanong, Varich","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087105600","display_name":"Dimitris Tsirogiannis","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tsirogiannis, Dimitris","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5111177928","display_name":"Anupam Datta","orcid":"https://orcid.org/0000-0001-9520-5196"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Datta, Anupam","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11719","display_name":"Data Quality and Management","score":0.23659999668598175,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11719","display_name":"Data Quality and Management","score":0.23659999668598175,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.19910000264644623,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.14790000021457672,"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/semantic-computing","display_name":"Semantic computing","score":0.5515000224113464},{"id":"https://openalex.org/keywords/semantic-compression","display_name":"Semantic compression","score":0.5315999984741211},{"id":"https://openalex.org/keywords/sql","display_name":"SQL","score":0.444599986076355},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4246000051498413},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.38929998874664307},{"id":"https://openalex.org/keywords/semantic-data-model","display_name":"Semantic data model","score":0.38339999318122864},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3499000072479248},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.3476000130176544},{"id":"https://openalex.org/keywords/semantic-web","display_name":"Semantic Web","score":0.3418000042438507}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8781999945640564},{"id":"https://openalex.org/C511149849","wikidata":"https://www.wikidata.org/wiki/Q7449051","display_name":"Semantic computing","level":3,"score":0.5515000224113464},{"id":"https://openalex.org/C202708506","wikidata":"https://www.wikidata.org/wiki/Q7449050","display_name":"Semantic compression","level":5,"score":0.5315999984741211},{"id":"https://openalex.org/C510870499","wikidata":"https://www.wikidata.org/wiki/Q47607","display_name":"SQL","level":2,"score":0.444599986076355},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43630000948905945},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4246000051498413},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.38929998874664307},{"id":"https://openalex.org/C90312973","wikidata":"https://www.wikidata.org/wiki/Q7449052","display_name":"Semantic data model","level":2,"score":0.38339999318122864},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3499000072479248},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.3476000130176544},{"id":"https://openalex.org/C2129575","wikidata":"https://www.wikidata.org/wiki/Q54837","display_name":"Semantic Web","level":2,"score":0.3418000042438507},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.33820000290870667},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3361000120639801},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33219999074935913},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.3212999999523163},{"id":"https://openalex.org/C85407183","wikidata":"https://www.wikidata.org/wiki/Q1045785","display_name":"Semantic network","level":2,"score":0.3174000084400177},{"id":"https://openalex.org/C103692084","wikidata":"https://www.wikidata.org/wiki/Q1765824","display_name":"Semantic grid","level":3,"score":0.313400000333786},{"id":"https://openalex.org/C168065819","wikidata":"https://www.wikidata.org/wiki/Q845566","display_name":"Debugging","level":2,"score":0.31209999322891235},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.311599999666214},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.31139999628067017},{"id":"https://openalex.org/C5655090","wikidata":"https://www.wikidata.org/wiki/Q192588","display_name":"Relational database","level":2,"score":0.3018999993801117},{"id":"https://openalex.org/C147497476","wikidata":"https://www.wikidata.org/wiki/Q54872","display_name":"RDF","level":3,"score":0.301800012588501},{"id":"https://openalex.org/C166423231","wikidata":"https://www.wikidata.org/wiki/Q1891170","display_name":"Semantic search","level":3,"score":0.3012000024318695},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29420000314712524},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C110903229","wikidata":"https://www.wikidata.org/wiki/Q7449064","display_name":"Semantic integration","level":4,"score":0.26510000228881836},{"id":"https://openalex.org/C196621874","wikidata":"https://www.wikidata.org/wiki/Q385210","display_name":"IDEF1X","level":4,"score":0.2623000144958496},{"id":"https://openalex.org/C113174947","wikidata":"https://www.wikidata.org/wiki/Q2859736","display_name":"Tree (set theory)","level":2,"score":0.2558000087738037},{"id":"https://openalex.org/C192028432","wikidata":"https://www.wikidata.org/wiki/Q845739","display_name":"Query language","level":2,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.07923","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.07923","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"doi:10.48550/arxiv.2606.07923","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.07923","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"score":0.637353777885437,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"With":[0],"the":[1,47,78,95,119,164],"advent":[2],"of":[3,80,98,121,220],"Large":[4],"Language":[5],"Models":[6],"(LLMs),":[7],"many":[8],"database":[9,54],"systems":[10],"introduced":[11],"semantic":[12,35,51,81,99,116,126,151,214],"operators":[13,26,100],"that":[14,226],"enabled":[15],"analytical":[16],"queries":[17,38,64],"over":[18],"unstructured":[19,110],"data":[20,111,132],"(e.g.":[21],"text,":[22],"images,":[23],"videos).":[24],"Semantic":[25],"typically":[27,113],"incur":[28],"high":[29,96],"inference":[30],"costs":[31],"and":[32,131,146,162,204,243],"latencies":[33],"making":[34,62],"(AI)":[36],"SQL":[37,85],"challenging":[39],"to":[40,56,66,181,189,241],"apply":[41],"on":[42,135],"large":[43],"scale":[44],"datasets.":[45],"At":[46],"same":[48],"time,":[49],"their":[50],"nature":[52],"leads":[53],"engines":[55],"treat":[57],"them":[58],"as":[59,168],"black":[60],"boxes,":[61],"AISQL":[63],"difficult":[65],"optimize.":[67],"In":[68,173],"this":[69],"paper,":[70],"we":[71,140],"introduce":[72],"Larch,":[73],"a":[74,169,177,191],"framework":[75],"for":[76,104,124,195],"optimizing":[77],"execution":[79],"filters":[82,152],"in":[83,118,218],"AI":[84],"queries.":[86],"Larch":[87,143,209,227],"was":[88],"inspired":[89],"by":[90,115,238],"two":[91,137,142],"key":[92,138],"observations:":[93],"i)":[94],"latency":[97],"leaves":[101],"significant":[102],"room":[103],"computationally-heavy":[105],"runtime":[106],"optimization":[107,216],"techniques,":[108],"ii)":[109],"are":[112],"accompanied":[114],"information":[117],"form":[120],"embeddings":[122],"allowing":[123],"efficient":[125],"comparisons":[127],"between":[128],"AI_FILTER":[129],"prompts":[130],"values.":[133],"Based":[134],"these":[136],"observations,":[139],"present":[141],"variants:":[144],"Larch-A2C":[145,148],"Larch-Sel.":[147],"encodes":[149],"arbitrary":[150],"expression":[153],"tree":[154],"using":[155],"an":[156],"embedding-augmented":[157],"Gated":[158],"Graph":[159],"Neural":[160],"Network":[161],"formulates":[163],"filter":[165,183,215],"evaluation":[166,193],"order":[167,194],"Markov":[170],"decision":[171],"process.":[172],"contrast,":[174],"Larch-Sel":[175],"leverages":[176],"supervised":[178],"learning":[179],"model":[180],"predict":[182],"selectivities,":[184],"subsequently":[185],"applying":[186],"dynamic":[187],"programming":[188],"find":[190],"near-optimal":[192],"each":[196],"input":[197],"row.":[198],"Evaluated":[199],"across":[200,230],"diverse":[201,231],"real-world":[202],"datasets":[203],"comprehensive":[205],"synthetic":[206],"workloads,":[207,232],"both":[208],"variants":[210],"always":[211],"outperform":[212],"existing":[213],"techniques":[217],"terms":[219],"token":[221,235],"usage.":[222],"Our":[223],"results":[224],"demonstrate":[225],"is":[228],"robust":[229],"reducing":[233],"total":[234],"cost":[236],"overhead":[237],"3x-19x":[239],"compared":[240],"Palimpzest":[242],"Quest.":[244]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-10T00:00:00"}