{"id":"https://openalex.org/W7139089707","doi":"https://doi.org/10.1609/aaai.v40i44.41064","title":"A Multi-Agent Conversational Bandit Approach to Online Evaluation and Selection of User-Aligned LLM Responses","display_name":"A Multi-Agent Conversational Bandit Approach to Online Evaluation and Selection of User-Aligned LLM Responses","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7139089707","doi":"https://doi.org/10.1609/aaai.v40i44.41064"},"language":"en","primary_location":{"id":"doi:10.1609/aaai.v40i44.41064","doi":"https://doi.org/10.1609/aaai.v40i44.41064","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i44.41064","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/41064/45025","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-2","norway-1"],"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://ojs.aaai.org/index.php/AAAI/article/download/41064/45025","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5079056866","display_name":"Xiangxiang Dai","orcid":"https://orcid.org/0000-0003-0179-196X","observed_orcids":["https://orcid.org/0000-0003-0179-196X"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xiangxiang Dai","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yuejin Xie","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuejin Xie","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073032248","display_name":"Maoli Liu","orcid":"https://orcid.org/0000-0002-6321-6576","observed_orcids":["https://orcid.org/0000-0002-6321-6576"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Maoli Liu","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069419940","display_name":"Xuchuang Wang","orcid":"https://orcid.org/0009-0006-8043-8521","observed_orcids":["https://orcid.org/0009-0006-8043-8521"]},"institutions":[{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xuchuang Wang","raw_affiliation_strings":["University of Massachusetts at Amherst"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Massachusetts at Amherst","institution_ids":["https://openalex.org/I24603500"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012872551","display_name":"Zhuohua Li","orcid":"https://orcid.org/0000-0002-1390-0222","observed_orcids":["https://orcid.org/0000-0002-1390-0222"]},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]},{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]},{"id":"https://openalex.org/I4405280064","display_name":"Guangzhou Institute of Technology","ror":"https://ror.org/02d1g9y70","country_code":"CN","type":"education","lineage":["https://openalex.org/I4405280064"]}],"countries":["CN","HK"],"is_corresponding":false,"raw_author_name":"Zhuohua Li","raw_affiliation_strings":["Guangzhou Institute of Technology, Xidian University\nThe Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Guangzhou Institute of Technology, Xidian University\nThe Chinese University of Hong Kong","institution_ids":["https://openalex.org/I149594827","https://openalex.org/I177725633","https://openalex.org/I4405280064"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130098771","display_name":"Huanyu Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huanyu Wang","raw_affiliation_strings":["Huawei Technologies Ltd"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Technologies Ltd","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5068489266","display_name":"John C. S. Lui","orcid":"https://orcid.org/0000-0001-7466-0384","observed_orcids":["https://orcid.org/0000-0001-7466-0384"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"John C.S. Lui","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]}],"institutions":[],"countries_distinct_count":3,"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":0.16799236,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"44","first_page":"37323","last_page":"37331"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9984999895095825,"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/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9984999895095825,"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/T10203","display_name":"Recommender Systems and Techniques","score":0.0006000000284984708,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":9.999999747378752e-05,"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/multi-agent-systems","display_name":"multi-agent systems","score":0.7300000190734863},{"id":"https://openalex.org/keywords/online-assessment","display_name":"online assessment","score":0.8100000023841858},{"id":"https://openalex.org/keywords/user-preference-alignment","display_name":"user preference alignment","score":0.9340000152587891},{"id":"https://openalex.org/keywords/regret-bounds","display_name":"regret bounds","score":0.9629999995231628}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8084999918937683},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.6398000121116638},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.5615000128746033},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5349000096321106},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5171999931335449},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.5047000050544739},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.47380000352859497},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.4641000032424927},{"id":"https://openalex.org/C50817715","wikidata":"https://www.wikidata.org/wiki/Q79895177","display_name":"Regret","level":2,"score":0.4577000141143799},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.45649999380111694},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.44690001010894775},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.362199991941452},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.34619998931884766},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.3199000060558319},{"id":"https://openalex.org/C2777851325","wikidata":"https://www.wikidata.org/wiki/Q7094102","display_name":"Online model","level":2,"score":0.3068000078201294},{"id":"https://openalex.org/C2778707766","wikidata":"https://www.wikidata.org/wiki/Q202064","display_name":"Phone","level":2,"score":0.3068000078201294},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.2856999933719635},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2721000015735626},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.265500009059906}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1609/aaai.v40i44.41064","doi":"https://doi.org/10.1609/aaai.v40i44.41064","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i44.41064","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/41064/45025","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-2","norway-1"],"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},{"id":"pmh:oai:ojs.aaai.org:article/41064","doi":null,"is_oa":false,"landing_page_url":"https://ojs.aaai.org/index.php/AAAI/article/view/41064","pdf_url":null,"source":{"id":"https://openalex.org/S4306400155","display_name":"AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI))","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2801652627","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/I2801652627"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2159-5399","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i44.41064","doi":"https://doi.org/10.1609/aaai.v40i44.41064","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i44.41064","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/41064/45025","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":["jufo-2","norway-1"],"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7139089707.pdf","grobid_xml":"https://content.openalex.org/works/W7139089707.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Prompt-based":[0],"offline":[1],"methods":[2,170],"are":[3],"commonly":[4],"used":[5],"to":[6,23,43,93,122],"optimize":[7],"large":[8,60],"language":[9],"model":[10,42],"(LLM)":[11],"responses,":[12],"but":[13],"evaluating":[14],"these":[15],"responses":[16,46],"is":[17],"computationally":[18],"intensive":[19],"and":[20,66,146,162,180],"often":[21],"fails":[22],"accommodate":[24],"diverse":[25],"response":[26,61,157,177],"styles.":[27],"This":[28],"study":[29],"introduces":[30],"a":[31,38],"novel":[32],"online":[33,90],"evaluation":[34],"framework":[35],"that":[36,129,165],"employs":[37,88],"multi-agent":[39],"conversational":[40,64],"bandit":[41],"select":[44],"optimal":[45],"while":[47],"aligning":[48],"with":[49,155,160],"user":[50],"preferences":[51],"dynamically.":[52],"To":[53],"tackle":[54],"challenges":[55],"such":[56],"as":[57],"high-dimensional":[58],"features,":[59],"sets,":[62],"adaptive":[63,116],"needs,":[65],"multi-device":[67],"access,":[68],"we":[69],"propose":[70],"MACO,":[71],"Multi-Agent":[72],"Conversational":[73],"Online":[74],"Learning,":[75],"which":[76],"comprises":[77],"two":[78],"key":[79],"components:":[80],"(1)":[81],"MACO-A:":[82],"Executed":[83,100],"by":[84,101,171],"local":[85],"agents,":[86],"it":[87,105],"an":[89],"elimination":[91],"mechanism":[92,118],"filter":[94],"out":[95],"low-quality":[96],"responses.":[97],"(2)":[98],"MACO-S:":[99],"the":[102,152],"cloud":[103],"server,":[104],"adaptively":[106],"adjusts":[107],"selection":[108],"strategies":[109],"based":[110],"on":[111,151],"aggregated":[112],"preference":[113,117],"data.":[114],"An":[115],"triggers":[119],"asynchronous":[120],"conversations":[121],"enhance":[123],"alignment":[124],"efficiency.":[125],"Theoretical":[126],"analysis":[127],"demonstrates":[128],"MACO":[130,166],"achieves":[131],"near-optimal":[132],"regret":[133],"bounds,":[134],"matching":[135],"state-of-the-art":[136],"performance":[137],"in":[138],"various":[139],"degenerate":[140],"cases.":[141],"Extensive":[142],"experiments":[143],"utilizing":[144],"Google":[145],"OpenAI":[147],"text":[148],"embedding":[149],"models":[150],"real-world":[153],"datasets":[154],"different":[156],"styles,":[158],"combined":[159],"Llama":[161],"GPT-4o,":[163],"show":[164],"consistently":[167],"outperforms":[168],"baseline":[169],"at":[172],"least":[173],"8.29%":[174],"across":[175],"varying":[176],"set":[178],"sizes":[179],"numbers":[181],"of":[182],"agents.":[183]},"counts_by_year":[],"updated_date":"2026-09-27T08:07:27.782135","created_date":"2026-03-20T00:00:00"}