{"id":"https://openalex.org/W7161978028","doi":"https://doi.org/10.1145/3786335.3813134","title":"Robust Batch-Level Query Routing for Large Language Models under Cost and Capacity Constraints","display_name":"Robust Batch-Level Query Routing for Large Language Models under Cost and Capacity Constraints","publication_year":2026,"publication_date":"2026-05-22","ids":{"openalex":"https://openalex.org/W7161978028","doi":"https://doi.org/10.1145/3786335.3813134"},"language":null,"primary_location":{"id":"doi:10.1145/3786335.3813134","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786335.3813134","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Conference on AI and Agentic Systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3786335.3813134","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101520719","display_name":"Jelena Markovi\u0107","orcid":"https://orcid.org/0000-0003-4962-3754"},"institutions":[{"id":"https://openalex.org/I1316064682","display_name":"LinkedIn (United States)","ror":"https://ror.org/02fyxhe35","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I1316064682"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jelena Markovic-Voronov","raw_affiliation_strings":["LinkedIn, Sunnyvale, CA, USA"],"raw_orcid":"https://orcid.org/0000-0002-0562-0386","affiliations":[{"raw_affiliation_string":"LinkedIn, Sunnyvale, CA, USA","institution_ids":["https://openalex.org/I1316064682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075423316","display_name":"Kayhan Behdin","orcid":"https://orcid.org/0000-0003-3482-0421"},"institutions":[{"id":"https://openalex.org/I1316064682","display_name":"LinkedIn (United States)","ror":"https://ror.org/02fyxhe35","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I1316064682"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kayhan Behdin","raw_affiliation_strings":["LinkedIn, Sunnyvale, CA, USA"],"raw_orcid":"https://orcid.org/0000-0003-3482-0421","affiliations":[{"raw_affiliation_string":"LinkedIn, Sunnyvale, CA, USA","institution_ids":["https://openalex.org/I1316064682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136648510","display_name":"Yuanda Xu","orcid":"https://orcid.org/0009-0008-9840-7883"},"institutions":[{"id":"https://openalex.org/I1316064682","display_name":"LinkedIn (United States)","ror":"https://ror.org/02fyxhe35","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I1316064682"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuanda Xu","raw_affiliation_strings":["LinkedIn, Sunnyvale, CA, USA"],"raw_orcid":"https://orcid.org/0009-0008-9840-7883","affiliations":[{"raw_affiliation_string":"LinkedIn, Sunnyvale, CA, USA","institution_ids":["https://openalex.org/I1316064682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136612070","display_name":"Zhengze Zhou","orcid":"https://orcid.org/0000-0002-5695-4386"},"institutions":[{"id":"https://openalex.org/I1316064682","display_name":"LinkedIn (United States)","ror":"https://ror.org/02fyxhe35","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I1316064682"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhengze Zhou","raw_affiliation_strings":["LinkedIn, Sunnyvale, CA, USA"],"raw_orcid":"https://orcid.org/0000-0002-5695-4386","affiliations":[{"raw_affiliation_string":"LinkedIn, Sunnyvale, CA, USA","institution_ids":["https://openalex.org/I1316064682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136616522","display_name":"Zhipeng Wang","orcid":"https://orcid.org/0000-0001-5703-5560"},"institutions":[{"id":"https://openalex.org/I1316064682","display_name":"LinkedIn (United States)","ror":"https://ror.org/02fyxhe35","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I1316064682"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhipeng Wang","raw_affiliation_strings":["LinkedIn, Sunnyvale, CA, USA"],"raw_orcid":"https://orcid.org/0000-0001-5703-5560","affiliations":[{"raw_affiliation_string":"LinkedIn, Sunnyvale, CA, USA","institution_ids":["https://openalex.org/I1316064682"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001765231","display_name":"Rahul Mazumder","orcid":"https://orcid.org/0000-0003-4285-7400"},"institutions":[{"id":"https://openalex.org/I1316064682","display_name":"LinkedIn (United States)","ror":"https://ror.org/02fyxhe35","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I1316064682"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rahul Mazumder","raw_affiliation_strings":["LinkedIn, Sunnyvale, CA, USA and Massachusetts Institute of Technology, Cambridge, MA, USA"],"raw_orcid":"https://orcid.org/0000-0003-1384-9743","affiliations":[{"raw_affiliation_string":"LinkedIn, Sunnyvale, CA, USA and Massachusetts Institute of Technology, Cambridge, MA, USA","institution_ids":["https://openalex.org/I1316064682"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.81576732,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"890","last_page":"916"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10064","display_name":"Complex Network Analysis Techniques","score":0.1573999971151352,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10064","display_name":"Complex Network Analysis Techniques","score":0.1573999971151352,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.09809999912977219,"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/T14347","display_name":"Big Data and Digital Economy","score":0.08780000358819962,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/routing","display_name":"Routing (electronic design automation)","score":0.4041999876499176},{"id":"https://openalex.org/keywords/query-language","display_name":"Query language","score":0.3928000032901764},{"id":"https://openalex.org/keywords/constraint","display_name":"Constraint (computer-aided design)","score":0.3359000086784363},{"id":"https://openalex.org/keywords/query-optimization","display_name":"Query optimization","score":0.322299987077713},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.29010000824928284},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.2897999882698059}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.685699999332428},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.4041999876499176},{"id":"https://openalex.org/C192028432","wikidata":"https://www.wikidata.org/wiki/Q845739","display_name":"Query language","level":2,"score":0.3928000032901764},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.3359000086784363},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.328900009393692},{"id":"https://openalex.org/C157692150","wikidata":"https://www.wikidata.org/wiki/Q2919848","display_name":"Query optimization","level":2,"score":0.322299987077713},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.31299999356269836},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2939000129699707},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.29010000824928284},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28859999775886536},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.27399998903274536},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2655999958515167},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.2615000009536743},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.25360000133514404}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3786335.3813134","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786335.3813134","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Conference on AI and Agentic Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3786335.3813134","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786335.3813134","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Conference on AI and Agentic Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W1647779468","https://openalex.org/W2295598076","https://openalex.org/W2970641574","https://openalex.org/W4244777963","https://openalex.org/W4251616545","https://openalex.org/W4387321091","https://openalex.org/W4387870549","https://openalex.org/W4394892775","https://openalex.org/W4411917733","https://openalex.org/W4415366360","https://openalex.org/W7117870265"],"related_works":[],"abstract_inverted_index":{"We":[0,60],"study":[1],"the":[2,107],"problem":[3],"of":[4,129],"routing":[5,21,43,111],"queries":[6],"to":[7,25,117,131,134],"large":[8],"language":[9],"models":[10],"(LLMs)":[11],"under":[12,30,119],"cost,":[13,28],"GPU":[14,144],"resources,":[15],"and":[16,56,84,122,143],"concurrency":[17],"constraints.":[18,146],"Prior":[19],"per-query":[20,113],"methods":[22,114],"often":[23],"fail":[24],"control":[26],"batch-level":[27,110],"especially":[29],"non-uniform":[31],"or":[32],"adversarial":[33,120],"batching.":[34],"To":[35],"address":[36],"this,":[37],"we":[38],"propose":[39],"a":[40,63,135],"batch-level,":[41],"resource-aware":[42],"framework":[44],"that":[45,66,81,96],"jointly":[46],"optimizes":[47],"model":[48,57],"assignment":[49],"for":[50,68],"each":[51],"batch":[52],"while":[53,139],"respecting":[54],"cost":[55,142],"capacity":[58],"limits.":[59],"further":[61],"introduce":[62],"robust":[64],"variant":[65],"accounts":[67],"uncertainty":[69],"in":[70],"predicted":[71],"LLM":[72,93],"performance,":[73],"along":[74],"with":[75],"an":[76],"offline":[77],"instance":[78,124],"allocation":[79,125],"procedure":[80],"balances":[82],"quality":[83],"throughput":[85],"across":[86],"multiple":[87],"models.":[88],"Experiments":[89],"on":[90,106],"two":[91],"multi-task":[92],"benchmarks":[94],"show":[95],"robustness":[97],"improves":[98],"accuracy":[99],"by":[100,115],"1\u201314%":[101],"over":[102],"non-robust":[103],"counterparts":[104],"(depending":[105],"performance":[108],"estimator),":[109],"outperforms":[112],"up":[116,130],"24%":[118],"batching,":[121],"optimized":[123],"yields":[126],"additional":[127],"gains":[128],"3%":[132],"compared":[133],"non-optimized":[136],"allocation,":[137],"all":[138],"strictly":[140],"controlling":[141],"resource":[145]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-22T00:00:00"}