{"id":"https://openalex.org/W4412887739","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1019","title":"Rethinking Diverse Human Preference Learning through Principal Component Analysis","display_name":"Rethinking Diverse Human Preference Learning through Principal Component Analysis","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412887739","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1019"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1019","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1019","pdf_url":"https://aclanthology.org/2025.findings-acl.1019.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: ACL 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.1019.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101542729","display_name":"Feng Luo","orcid":"https://orcid.org/0000-0003-0615-4695"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feng Luo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017503364","display_name":"Rui Yang","orcid":"https://orcid.org/0000-0001-5695-3773"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rui Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100359494","display_name":"Hao Sun","orcid":"https://orcid.org/0000-0003-4873-6992"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080098681","display_name":"Chunyuan Deng","orcid":"https://orcid.org/0000-0002-3537-493X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chunyuan Deng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111274045","display_name":"Jiarui Yao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiarui Yao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048753906","display_name":"Jingyan Shen","orcid":"https://orcid.org/0000-0003-4609-0251"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jingyan Shen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100356942","display_name":"Huan Zhang","orcid":"https://orcid.org/0000-0001-9685-2803"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huan Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5036933610","display_name":"Hanjie Chen","orcid":"https://orcid.org/0000-0001-6577-4830"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hanjie Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"19857","last_page":"19870"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12496","display_name":"Color perception and design","score":0.5819000005722046,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12496","display_name":"Color perception and design","score":0.5819000005722046,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/principal-component-analysis","display_name":"Principal component analysis","score":0.7655762434005737},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5935602784156799},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.5722155570983887},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.5633270740509033},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4618811309337616},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.15958786010742188},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.14905661344528198},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.05327713489532471}],"concepts":[{"id":"https://openalex.org/C27438332","wikidata":"https://www.wikidata.org/wiki/Q2873","display_name":"Principal component analysis","level":2,"score":0.7655762434005737},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5935602784156799},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.5722155570983887},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5633270740509033},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4618811309337616},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.15958786010742188},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.14905661344528198},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.05327713489532471},{"id":"https://openalex.org/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1019","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1019","pdf_url":"https://aclanthology.org/2025.findings-acl.1019.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: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.1019","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1019","pdf_url":"https://aclanthology.org/2025.findings-acl.1019.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: ACL 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/W4412887739.pdf","grobid_xml":"https://content.openalex.org/works/W4412887739.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W1975632186","https://openalex.org/W3027745756","https://openalex.org/W3205213561","https://openalex.org/W2531880140","https://openalex.org/W2763148304","https://openalex.org/W2036609560","https://openalex.org/W346861917"],"abstract_inverted_index":{"Understanding":[0],"human":[1,59,73],"preferences":[2,14,60,74],"is":[3,39,70],"crucial":[4],"for":[5,23,156],"improving":[6],"foundation":[7],"models":[8,26],"and":[9,18,41,77,93,122,141,158],"building":[10],"personalized":[11,157],"AI":[12],"systems.However,":[13],"are":[15],"inherently":[16],"diverse":[17,58],"complex,":[19],"making":[20],"it":[21,38],"difficult":[22],"traditional":[24,126],"reward":[25,127],"to":[27,43,71,113,125,143],"capture":[28,102],"their":[29],"full":[30],"range.While":[31],"fine-grained":[32,66],"preference":[33,135],"data":[34],"can":[35,109],"help,":[36],"collecting":[37],"expensive":[40],"hard":[42],"scale.In":[44],"this":[45],"paper,":[46],"we":[47],"introduce":[48],"Decomposed":[49],"Reward":[50],"Models":[51],"(DRMs),":[52],"a":[53,86,153],"novel":[54],"approach":[55],"that":[56,101,130],"extracts":[57],"from":[61],"binary":[62],"comparisons":[63],"without":[64,146],"requiring":[65],"annotations.Our":[67],"key":[68],"insight":[69],"represent":[72],"as":[75,152],"vectors":[76,100],"analyze":[78],"them":[79],"using":[80],"Principal":[81],"Component":[82],"Analysis":[83],"(PCA).By":[84],"constructing":[85],"dataset":[87],"of":[88,105],"embedding":[89],"differences":[90],"between":[91],"preferred":[92],"rejected":[94],"responses,":[95],"DRMs":[96,131,151],"identify":[97],"orthogonal":[98],"basis":[99],"distinct":[103],"aspects":[104],"preference.These":[106],"decomposed":[107],"rewards":[108],"be":[110],"flexibly":[111],"combined":[112],"align":[114],"with":[115],"different":[116],"user":[117],"needs,":[118],"offering":[119],"an":[120],"interpretable":[121,159],"scalable":[123],"alternative":[124],"models.We":[128],"demonstrate":[129],"effectively":[132],"extract":[133],"meaningful":[134],"dimensions":[136],"(e.g.,":[137],"helpfulness,":[138],"safety,":[139],"humor)":[140],"adapt":[142],"new":[144],"users":[145],"additional":[147],"training.Our":[148],"results":[149],"highlight":[150],"powerful":[154],"framework":[155],"LLM":[160],"alignment.":[161]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}