{"id":"https://openalex.org/W4414004064","doi":"https://doi.org/10.14778/3749646.3749702","title":"ThriftLLM: On Cost-Effective Selection of Large Language Models for Classification Queries","display_name":"ThriftLLM: On Cost-Effective Selection of Large Language Models for Classification Queries","publication_year":2025,"publication_date":"2025-07-01","ids":{"openalex":"https://openalex.org/W4414004064","doi":"https://doi.org/10.14778/3749646.3749702"},"language":"en","primary_location":{"id":"doi:10.14778/3749646.3749702","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3749646.3749702","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-article"},"type":"article","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/A5082518323","display_name":"Keke Huang","orcid":"https://orcid.org/0000-0003-2190-7114"},"institutions":[{"id":"https://openalex.org/I141945490","display_name":"University of British Columbia","ror":"https://ror.org/03rmrcq20","country_code":"CA","type":"education","lineage":["https://openalex.org/I141945490"]}],"countries":["CA"],"is_corresponding":true,"raw_author_name":"Keke Huang","raw_affiliation_strings":["University of British Columbia"],"affiliations":[{"raw_affiliation_string":"University of British Columbia","institution_ids":["https://openalex.org/I141945490"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102863666","display_name":"Yimin Shi","orcid":"https://orcid.org/0000-0003-0940-2797"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yimin Shi","raw_affiliation_strings":["National University of Singapore"],"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085340125","display_name":"Dujian Ding","orcid":null},"institutions":[{"id":"https://openalex.org/I141945490","display_name":"University of British Columbia","ror":"https://ror.org/03rmrcq20","country_code":"CA","type":"education","lineage":["https://openalex.org/I141945490"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Dujian Ding","raw_affiliation_strings":["University of British Columbia"],"affiliations":[{"raw_affiliation_string":"University of British Columbia","institution_ids":["https://openalex.org/I141945490"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100317444","display_name":"Yifei Li","orcid":"https://orcid.org/0009-0000-3562-4815"},"institutions":[{"id":"https://openalex.org/I141945490","display_name":"University of British Columbia","ror":"https://ror.org/03rmrcq20","country_code":"CA","type":"education","lineage":["https://openalex.org/I141945490"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Yifei Li","raw_affiliation_strings":["University of British Columbia"],"affiliations":[{"raw_affiliation_string":"University of British Columbia","institution_ids":["https://openalex.org/I141945490"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084674829","display_name":"Fei Yang","orcid":"https://orcid.org/0000-0002-2242-7158"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yang Fei","raw_affiliation_strings":["National University of Singapore"],"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061340195","display_name":"Laks V. S. Lakshmanan","orcid":"https://orcid.org/0000-0002-9775-4241"},"institutions":[{"id":"https://openalex.org/I141945490","display_name":"University of British Columbia","ror":"https://ror.org/03rmrcq20","country_code":"CA","type":"education","lineage":["https://openalex.org/I141945490"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Laks Lakshmanan","raw_affiliation_strings":["University of British Columbia"],"affiliations":[{"raw_affiliation_string":"University of British Columbia","institution_ids":["https://openalex.org/I141945490"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5010903591","display_name":"Xiaokui Xiao","orcid":"https://orcid.org/0000-0003-0914-4580"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Xiaokui Xiao","raw_affiliation_strings":["National University of Singapore, CNRS@CREATE, Singapore"],"affiliations":[{"raw_affiliation_string":"National University of Singapore, CNRS@CREATE, Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5082518323"],"corresponding_institution_ids":["https://openalex.org/I141945490"],"apc_list":null,"apc_paid":null,"fwci":4.4044,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.94986356,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":"18","issue":"11","first_page":"4410","last_page":"4423"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11719","display_name":"Data Quality and Management","score":0.9994000196456909,"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":0.9994000196456909,"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/T10028","display_name":"Topic Modeling","score":0.9987000226974487,"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":0.996399998664856,"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/selection","display_name":"Selection (genetic algorithm)","score":0.7718340754508972},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.710162878036499},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4893169105052948},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4342382550239563},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.38233625888824463}],"concepts":[{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.7718340754508972},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.710162878036499},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4893169105052948},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4342382550239563},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.38233625888824463}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.14778/3749646.3749702","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3749646.3749702","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1776713878","https://openalex.org/W2080379754","https://openalex.org/W2092799168","https://openalex.org/W2132801025","https://openalex.org/W2158557760","https://openalex.org/W2592313125","https://openalex.org/W2798649495","https://openalex.org/W2946609015","https://openalex.org/W2971681342","https://openalex.org/W3029654587","https://openalex.org/W3045492832","https://openalex.org/W3103177583","https://openalex.org/W3123375411","https://openalex.org/W3155638005","https://openalex.org/W4233413206","https://openalex.org/W4246012463","https://openalex.org/W4254362479","https://openalex.org/W4281721601","https://openalex.org/W4385571189","https://openalex.org/W4391188659","https://openalex.org/W4391494845","https://openalex.org/W4396571402","https://openalex.org/W4396757505","https://openalex.org/W4398234583","https://openalex.org/W4400529373","https://openalex.org/W4402043024","https://openalex.org/W4402043038","https://openalex.org/W4402043304","https://openalex.org/W4404181042","https://openalex.org/W4404181300","https://openalex.org/W4404407667","https://openalex.org/W4404792811"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Recently,":[0],"large":[1],"language":[2,14],"models":[3,125],"(LLMs)":[4,126],"have":[5],"demonstrated":[6],"remarkable":[7],"capabilities":[8],"in":[9,19,40,267],"understanding":[10],"and":[11,22,55,188,190,217,256,283],"generating":[12],"natural":[13],"content,":[15],"attracting":[16],"widespread":[17],"attention":[18],"both":[20],"industry":[21],"academia.":[23],"An":[24],"increasing":[25],"number":[26],"of":[27,43,51,72,89,121,124,130,162,166,285],"services":[28],"offer":[29],"LLMs":[30,37,50,146,167],"for":[31,68,141,253],"various":[32,265],"tasks":[33],"via":[34],"APIs.":[35],"Different":[36],"demonstrate":[38],"expertise":[39],"different":[41,52],"domains":[42],"queries":[44,259],"(e.g.,":[45],"text":[46,254],"classification":[47,255],"queries).":[48],"Meanwhile,":[49],"scales,":[53],"complexities,":[54],"performance":[56,111,120,252],"are":[57,65],"priced":[58],"diversely.":[59],"Driven":[60],"by":[61],"this,":[62,153],"several":[63],"researchers":[64],"investigating":[66],"strategies":[67],"selecting":[69,163],"an":[70,99,122,139,214,222],"ensemble":[71,101,110,123,149],"LLMs,":[73],"aiming":[74],"to":[75,84,97,103,147,169,199,242],"decrease":[76],"overall":[77,176],"usage":[78],"costs":[79],"while":[80,272],"enhancing":[81],"performance.":[82,150],"However,":[83],"our":[85,268,286],"best":[86],"knowledge,":[87],"none":[88],"the":[90,94,109,119,128,156,175,182,194,281],"existing":[91],"works":[92],"addresses":[93],"problem,":[95],"how":[96],"find":[98],"LLM":[100,240],"subject":[102,168],"a":[104,164,170,204,233,274],"cost":[105,171,277],"budget,":[106,278],"which":[107,133],"maximizes":[108,174],"with":[112,226],"guarantees.":[113],"In":[114],"this":[115],"paper,":[116],"we":[117,134,154,212],"formalize":[118],"using":[127,273],"notion":[129],"correctness":[131,177,183,210],"probability,":[132,211],"formally":[135],"define.":[136],"We":[137,179],"develop":[138,213],"approach":[140],"aggregating":[142],"responses":[143],"from":[144],"multiple":[145,261],"enhance":[148],"Building":[151],"on":[152,260],"formulate":[155],"Optimal":[157],"Ensemble":[158],"Selection":[159],"(OES)":[160],"problem":[161,196],"set":[165],"budget":[172,247],"that":[173,181,193,207,219,237],"probability.":[178,228],"show":[180],"probability":[184],"function":[185,206],"is":[186,197],"non-decreasing":[187],"non-submodular":[189],"provide":[191],"evidence":[192],"OES":[195],"likely":[198],"be":[200],"NP-hard.":[201],"By":[202],"leveraging":[203],"submodular":[205],"upper":[208],"bounds":[209],"algorithm,":[215],"ThriftLLM,":[216],"prove":[218],"it":[220],"achieves":[221,250],"instance-dependent":[223],"approximation":[224],"guarantee":[225],"high":[227],"Our":[229],"framework":[230],"functions":[231],"as":[232],"data":[234],"processing":[235],"system":[236],"selects":[238],"appropriate":[239],"operators":[241],"deliver":[243],"high-quality":[244],"results":[245],"under":[246],"constraints.":[248],"It":[249],"state-of-the-art":[251],"entity":[257],"matching":[258],"real-world":[262],"datasets":[263],"against":[264],"baselines":[266],"extensive":[269],"experimental":[270],"evaluation,":[271],"relatively":[275],"lower":[276],"strongly":[279],"supporting":[280],"effectiveness":[282],"superiority":[284],"method.":[287]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}