{"id":"https://openalex.org/W7137866476","doi":"https://doi.org/10.1609/aaai.v40i31.39815","title":"Reward Model Evaluation via Automatically-Ranked Policy Alignment","display_name":"Reward Model Evaluation via Automatically-Ranked Policy Alignment","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7137866476","doi":"https://doi.org/10.1609/aaai.v40i31.39815"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i31.39815","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i31.39815","pdf_url":null,"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":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i31.39815","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129736649","display_name":"Aoran Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Aoran Wang","raw_affiliation_strings":["Nanjing University"],"affiliations":[{"raw_affiliation_string":"Nanjing University","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129725137","display_name":"Lei Ou","orcid":null},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei Ou","raw_affiliation_strings":["Nanjing University"],"affiliations":[{"raw_affiliation_string":"Nanjing University","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129646568","display_name":"Yang YU","orcid":null},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yang Yu","raw_affiliation_strings":["Nanjing University"],"affiliations":[{"raw_affiliation_string":"Nanjing University","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5023151878","display_name":"ZongZhang Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zongzhang Zhang","raw_affiliation_strings":["Nanjing University"],"affiliations":[{"raw_affiliation_string":"Nanjing University","institution_ids":["https://openalex.org/I881766915"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5129736649"],"corresponding_institution_ids":["https://openalex.org/I881766915"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.08029979,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"31","first_page":"26124","last_page":"26132"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.5426999926567078,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.5426999926567078,"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":0.061500001698732376,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.04490000009536743,"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":0.6588000059127808},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6276000142097473},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.6115999817848206},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.532800018787384},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.501800000667572},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.4603999853134155}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7210000157356262},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6588000059127808},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6276000142097473},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.6115999817848206},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.532800018787384},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.501800000667572},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.4603999853134155},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.44670000672340393},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44510000944137573},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4140999913215637},{"id":"https://openalex.org/C2780898871","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Performance metric","level":2,"score":0.3930000066757202},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.31279999017715454},{"id":"https://openalex.org/C143661069","wikidata":"https://www.wikidata.org/wiki/Q670713","display_name":"Reward system","level":2,"score":0.3009999990463257},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.29580000042915344},{"id":"https://openalex.org/C2779110102","wikidata":"https://www.wikidata.org/wiki/Q1323737","display_name":"Revealed preference","level":2,"score":0.2720000147819519}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i31.39815","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i31.39815","pdf_url":null,"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"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i31.39815","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i31.39815","pdf_url":null,"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":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Evaluating":[0],"reward":[1,16,41,62,91,108,177,204],"models":[2,92],"is":[3,18,71],"a":[4,106,116,131,147],"fundamental":[5],"challenge":[6],"in":[7,12,115,158,202],"Reinforcement":[8],"Learning":[9],"(RL),":[10],"particularly":[11],"settings":[13],"where":[14],"the":[15,39,48,51,58,61,65,76,125,159,165,175],"model":[17,42,45,63,205],"learned":[19],"or":[20,109],"manually":[21],"designed.":[22],"The":[23],"standard":[24],"paradigm":[25],"for":[26,78,134],"Reward":[27],"Model":[28],"Evaluation":[29],"(RME)":[30],"involves":[31],"training":[32,140],"an":[33],"optimal":[34],"policy":[35,79,152,160,196],"via":[36],"RL":[37,68,139],"on":[38,99],"given":[40],"and":[43,70,162,171,183,198],"assessing":[44],"quality":[46,59],"through":[47],"performance":[49],"of":[50,60,67,149],"resulting":[52],"policy.":[53],"However,":[54],"this":[55,87],"approach":[56],"conflates":[57],"with":[64,193],"effectiveness":[66],"training,":[69],"computationally":[72],"expensive":[73],"due":[74],"to":[75,85,105,111],"need":[77],"optimization.":[80],"Recent":[81],"RME":[82,135],"methods":[83],"attempt":[84],"circumvent":[86],"issue":[88],"by":[89,174],"evaluating":[90,203],"directly,":[93],"without":[94],"RL,":[95],"but":[96],"often":[97],"rely":[98],"impractical":[100],"assumptions":[101],"such":[102],"as":[103],"access":[104],"ground-truth":[107,142],"fail":[110],"utilize":[112],"available":[113],"supervision":[114],"fine-grained":[117],"manner.":[118],"To":[119],"overcome":[120],"these":[121,168],"limitations,":[122],"we":[123],"propose":[124],"Policy":[126],"Preference":[127],"Alignment":[128],"Coefficient":[129],"(PPAC),":[130],"novel":[132],"metric":[133],"that":[136,154,188],"requires":[137],"neither":[138],"nor":[141],"rewards.":[143],"PPAC":[144,189],"first":[145],"generates":[146],"sequence":[148],"automatically":[150],"ranked":[151],"preferences":[153,170],"guarantee":[155],"monotonic":[156],"improvement":[157],"value,":[161],"then":[163],"quantifies":[164],"alignment":[166],"between":[167],"generated":[169],"those":[172],"implied":[173],"candidate":[176],"model.":[178],"Experimental":[179],"results":[180],"across":[181],"gridworld":[182],"continuous":[184],"control":[185],"task":[186],"demonstrate":[187],"yields":[190],"preference":[191],"sequences":[192],"consistently":[194],"increasing":[195],"values":[197],"outperforms":[199],"existing":[200],"metrics":[201],"quality.":[206]},"counts_by_year":[],"updated_date":"2026-03-20T20:47:17.329874","created_date":"2026-03-18T00:00:00"}