{"id":"https://openalex.org/W4416034518","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1301","title":"Adaptive LLM Routing under Budget Constraints","display_name":"Adaptive LLM Routing under Budget Constraints","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034518","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1301"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1301","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1301","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1301.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.1301.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5111226571","display_name":"Pranoy Panda","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pranoy Panda","raw_affiliation_strings":["Fujitsu Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fujitsu Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120292014","display_name":"Raghav Magazine","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Raghav Magazine","raw_affiliation_strings":["Fujitsu Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fujitsu Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099096845","display_name":"Chaitanya Devaguptapu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chaitanya Devaguptapu","raw_affiliation_strings":["Fujitsu Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fujitsu Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074964546","display_name":"Sho Takemori","orcid":"https://orcid.org/0000-0003-2303-824X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sho Takemori","raw_affiliation_strings":["Fujitsu Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fujitsu Research","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5037469337","display_name":"Vishal Sharma","orcid":"https://orcid.org/0000-0003-3219-8958"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vishal Sharma","raw_affiliation_strings":["Fujitsu Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fujitsu Research","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.28353104,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"23934","last_page":"23949"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10847","display_name":"Advanced Optical Network Technologies","score":0.47940000891685486,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10847","display_name":"Advanced Optical Network Technologies","score":0.47940000891685486,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10138","display_name":"Network Traffic and Congestion Control","score":0.07649999856948853,"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/T10714","display_name":"Software-Defined Networks and 5G","score":0.046799998730421066,"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/embedding","display_name":"Embedding","score":0.6459000110626221},{"id":"https://openalex.org/keywords/routing","display_name":"Routing (electronic design automation)","score":0.5997999906539917},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.54830002784729},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.4090000092983246},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.3889999985694885},{"id":"https://openalex.org/keywords/online-learning","display_name":"Online learning","score":0.3596000075340271},{"id":"https://openalex.org/keywords/knapsack-problem","display_name":"Knapsack problem","score":0.3513999879360199},{"id":"https://openalex.org/keywords/dual","display_name":"Dual (grammatical number)","score":0.3395000100135803}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7301999926567078},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6459000110626221},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.5997999906539917},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.54830002784729},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.4090000092983246},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39089998602867126},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.3889999985694885},{"id":"https://openalex.org/C2986087404","wikidata":"https://www.wikidata.org/wiki/Q15946010","display_name":"Online learning","level":2,"score":0.3596000075340271},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35929998755455017},{"id":"https://openalex.org/C113138325","wikidata":"https://www.wikidata.org/wiki/Q864457","display_name":"Knapsack problem","level":2,"score":0.3513999879360199},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.3395000100135803},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.3199000060558319},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.3174999952316284},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.30959999561309814},{"id":"https://openalex.org/C8505890","wikidata":"https://www.wikidata.org/wiki/Q605095","display_name":"Budget constraint","level":2,"score":0.28859999775886536},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.2833999991416931},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.2797999978065491},{"id":"https://openalex.org/C2778029271","wikidata":"https://www.wikidata.org/wiki/Q5421931","display_name":"Extension (predicate logic)","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.27549999952316284},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2538999915122986},{"id":"https://openalex.org/C204948658","wikidata":"https://www.wikidata.org/wiki/Q1119410","display_name":"Static routing","level":4,"score":0.25369998812675476}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.1301","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1301","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1301.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1301","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1301","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1301.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034518.pdf","grobid_xml":"https://content.openalex.org/works/W4416034518.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"revolutionized":[5],"natural":[6],"language":[7],"processing,":[8],"but":[9],"their":[10,113],"varying":[11],"capabilities":[12],"and":[13,54,102,106,124],"costs":[14],"pose":[15],"challenges":[16],"in":[17],"practical":[18],"applications.LLM":[19],"routing":[20,64],"addresses":[21],"this":[22,35,92,131],"by":[23],"dynamically":[24],"selecting":[25],"the":[26],"most":[27],"suitable":[28],"LLM":[29,63,107],"for":[30,83,100,150],"each":[31],"query/task.Previous":[32],"approaches":[33],"treat":[34],"as":[36,65,160],"a":[37,66,96,141,161],"supervised":[38,89],"learning":[39],"problem,":[40,69,93,164],"assuming":[41],"complete":[42],"knowledge":[43],"of":[44,144],"optimal":[45],"query-LLM":[46],"pairings.However,":[47],"realworld":[48],"scenarios":[49],"lack":[50],"such":[51],"comprehensive":[52],"mappings":[53],"face":[55],"evolving":[56],"user":[57,148],"queries.We":[58],"thus":[59],"propose":[60],"to":[61,88,111],"study":[62],"contextual":[67],"bandit":[68,74,128],"enabling":[70],"adaptive":[71],"decisionmaking":[72],"using":[73],"feedback":[75],"without":[76],"requiring":[77],"exhaustive":[78],"inference":[79],"across":[80],"all":[81,84],"LLMs":[82],"queries":[85,101],"(in":[86],"contrast":[87],"routing).To":[90],"address":[91],"we":[94,153],"develop":[95],"shared":[97],"embedding":[98],"space":[99,115],"LLMs,":[103],"where":[104],"query":[105],"embeddings":[108],"are":[109],"aligned":[110],"reflect":[112],"affinity.This":[114],"is":[116],"initially":[117],"learned":[118],"from":[119],"offline":[120],"human":[121],"preference":[122],"data":[123],"refined":[125],"through":[126,133],"online":[127,156],"feedback.We":[129],"instantiate":[130],"idea":[132],"Preference-prior":[134],"Informed":[135],"LinUCB":[136],"fOr":[137],"Adaptive":[138],"RouTing":[139],"(PILOT),":[140],"novel":[142],"extension":[143],"LinUCB.To":[145],"handle":[146],"diverse":[147],"budgets":[149],"model":[151],"routing,":[152],"introduce":[154],"an":[155],"cost":[157],"policy":[158],"modeled":[159],"multi-choice":[162],"knapsack":[163],"ensuring":[165],"resource-efficient":[166],"routing.":[167]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}