{"id":"https://openalex.org/W7139111931","doi":"https://doi.org/10.1609/aaai.v40i44.41110","title":"SharedRep-RLHF: A Shared Representation Approach to RLHF with Diverse Preferences","display_name":"SharedRep-RLHF: A Shared Representation Approach to RLHF with Diverse Preferences","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7139111931","doi":"https://doi.org/10.1609/aaai.v40i44.41110"},"language":"en","primary_location":{"id":"doi:10.1609/aaai.v40i44.41110","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i44.41110","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/41110/45071","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,"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"},"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/41110/45071","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5126478040","display_name":"Arpan Mukherjee","orcid":null},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Arpan Mukherjee","raw_affiliation_strings":["Imperial College London"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Imperial College London","institution_ids":["https://openalex.org/I47508984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120650810","display_name":"Marcello Bullo","orcid":null},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Marcello Bullo","raw_affiliation_strings":["Imperial College London"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Imperial College London","institution_ids":["https://openalex.org/I47508984"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5130006682","display_name":"Deniz G\u00fcnd\u00fcz","orcid":null},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Deniz G\u00fcnd\u00fcz","raw_affiliation_strings":["Imperial College London"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Imperial College London","institution_ids":["https://openalex.org/I47508984"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I47508984"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.22761787,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"44","first_page":"37747","last_page":"37755"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.2858000099658966,"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"}},"topics":[{"id":"https://openalex.org/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.2858000099658966,"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/T10667","display_name":"Emotion and Mood Recognition","score":0.13349999487400055,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.0877000018954277,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.6389999985694885},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.5382999777793884},{"id":"https://openalex.org/keywords/contrast","display_name":"Contrast (vision)","score":0.527899980545044},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5192000269889832},{"id":"https://openalex.org/keywords/preference-learning","display_name":"Preference learning","score":0.5087000131607056},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5062000155448914},{"id":"https://openalex.org/keywords/diversity","display_name":"Diversity (politics)","score":0.44909998774528503}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7117999792098999},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6389999985694885},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5953999757766724},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5382999777793884},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.527899980545044},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5192000269889832},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.5087000131607056},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5062000155448914},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4975999891757965},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.44909998774528503},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.3813000023365021},{"id":"https://openalex.org/C2778445095","wikidata":"https://www.wikidata.org/wiki/Q18354077","display_name":"Sample complexity","level":2,"score":0.36059999465942383},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.3560999929904938},{"id":"https://openalex.org/C2781311116","wikidata":"https://www.wikidata.org/wiki/Q83306","display_name":"Group (periodic table)","level":2,"score":0.34119999408721924},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.3334999978542328},{"id":"https://openalex.org/C2776608160","wikidata":"https://www.wikidata.org/wiki/Q4785462","display_name":"Natural (archaeology)","level":2,"score":0.3303000032901764},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.3082999885082245},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.30219998955726624}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1609/aaai.v40i44.41110","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i44.41110","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/41110/45071","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,"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/41110","is_oa":false,"landing_page_url":"https://ojs.aaai.org/index.php/AAAI/article/view/41110","pdf_url":null,"source":{"id":"https://openalex.org/S7407064318","display_name":"AAAI Publications","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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.v40i44.41110","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i44.41110","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/41110/45071","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,"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":[],"awards":[{"id":"https://openalex.org/G2151786554","display_name":null,"funder_award_id":"EP/X030806/1","funder_id":"https://openalex.org/F4320314731","funder_display_name":"UK Research and Innovation"},{"id":"https://openalex.org/G759609298","display_name":null,"funder_award_id":"EP/Y028732/1","funder_id":"https://openalex.org/F4320314731","funder_display_name":"UK Research and Innovation"}],"funders":[{"id":"https://openalex.org/F4320314731","display_name":"UK Research and Innovation","ror":"https://ror.org/001aqnf71"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7139111931.pdf","grobid_xml":"https://content.openalex.org/works/W7139111931.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Uniform-reward":[0],"reinforcement":[1],"learning":[2,39,105,120],"from":[3],"human":[4],"feedback":[5],"(RLHF),":[6],"which":[7],"trains":[8],"a":[9,60,74,82,147],"single":[10],"reward":[11,41,107],"model":[12],"to":[13,21,104,144,151],"represent":[14],"the":[15,23,47,50,70,126,139],"preferences":[16],"of":[17,25,63,129,141,149],"all":[18],"annotators,":[19],"fails":[20],"capture":[22],"diversity":[24],"opinions":[26],"across":[27,109,133],"sub-populations,":[28],"inadvertently":[29],"favoring":[30],"dominant":[31],"groups.":[32,110],"The":[33],"state-of-the-art,":[34],"MaxMin-RLHF,":[35],"addresses":[36],"this":[37,78],"by":[38,44],"group-specific":[40],"models,":[42],"and":[43,92,123],"optimizing":[45],"for":[46],"group":[48,72],"receiving":[49],"minimum":[51],"reward,":[52],"thereby":[53],"promoting":[54],"fairness.":[55],"However,":[56],"we":[57,80],"identify":[58],"that":[59,114],"key":[61],"limitation":[62],"MaxMin-RLHF":[64,115,145],"is":[65,73,116],"its":[66,88],"poor":[67],"performance":[68],"when":[69],"minimum-reward":[71],"minority.":[75],"To":[76],"mitigate":[77],"drawback,":[79],"introduce":[81],"novel":[83],"framework,":[84],"termed":[85],"*SharedRep-RLHF*.":[86],"At":[87],"core,":[89],"SharedRep-RLHF":[90],"learns":[91],"leverages":[93],"*shared":[94],"preference":[95],"traits*":[96],"in":[97,102,119,153],"annotations":[98],"among":[99],"various":[100],"groups,":[101],"contrast":[103],"separate":[106],"models":[108],"We":[111],"first":[112],"show":[113],"provably":[117],"suboptimal":[118],"shared":[121],"traits,":[122],"then":[124],"quantify":[125],"sample":[127],"complexity":[128],"SharedRep":[130],"RLHF.":[131],"Experiments":[132],"diverse":[134],"natural":[135],"language":[136],"tasks":[137],"showcase":[138],"effectiveness":[140],"ShareRep-RLHF":[142],"compared":[143],"with":[146],"gain":[148],"up":[150],"20%":[152],"win":[154],"rate.":[155]},"counts_by_year":[],"updated_date":"2026-09-02T07:38:04.252551","created_date":"2026-03-20T00:00:00"}