{"id":"https://openalex.org/W4412944928","doi":"https://doi.org/10.18653/v1/2025.acl-long.946","title":"From Outcomes to Processes: Guiding PRM Learning from ORM for Inference-Time Alignment","display_name":"From Outcomes to Processes: Guiding PRM Learning from ORM for Inference-Time Alignment","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412944928","doi":"https://doi.org/10.18653/v1/2025.acl-long.946"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.946","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.946","pdf_url":"https://aclanthology.org/2025.acl-long.946.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.946.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103125014","display_name":"Bin Xie","orcid":"https://orcid.org/0000-0001-5118-3570"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bin Xie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031468255","display_name":"Bingbing Xu","orcid":"https://orcid.org/0009-0004-8319-2681"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bingbing Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101538733","display_name":"Yige Yuan","orcid":"https://orcid.org/0000-0001-8856-668X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yige Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108336134","display_name":"S. Zhu","orcid":"https://orcid.org/0009-0006-0223-9092"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shengmao Zhu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5103007496","display_name":"Huawei Shen","orcid":"https://orcid.org/0000-0003-1204-4820"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huawei Shen","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":"19291","last_page":"19307"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12205","display_name":"Time Series Analysis and Forecasting","score":0.9271000027656555,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T12205","display_name":"Time Series Analysis and Forecasting","score":0.9271000027656555,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T11719","display_name":"Data Quality and Management","score":0.9050999879837036,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7629250288009644},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7011754512786865},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.49796223640441895},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.40830671787261963}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7629250288009644},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7011754512786865},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49796223640441895},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40830671787261963}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.946","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.946","pdf_url":"https://aclanthology.org/2025.acl-long.946.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.946","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.946","pdf_url":"https://aclanthology.org/2025.acl-long.946.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G355116499","display_name":null,"funder_award_id":"62202448","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G740642855","display_name":null,"funder_award_id":"U21B2046","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412944928.pdf","grobid_xml":"https://content.openalex.org/works/W4412944928.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W4387369504","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"Inference-time":[0],"alignment":[1],"methods":[2,54],"have":[3],"gained":[4],"significant":[5],"attention":[6],"for":[7,49],"their":[8],"efficiency":[9],"and":[10,67,81,98,101,123,138],"effectiveness":[11],"in":[12,152],"aligning":[13,104],"large":[14],"language":[15],"models":[16,33,77],"(LLMs)":[17],"with":[18,108],"human":[19,109,132],"preferences.However,":[20],"existing":[21,146],"dominant":[22],"approaches":[23],"using":[24],"rewardguided":[25],"search":[26],"(RGS)":[27],"primarily":[28],"rely":[29,55],"on":[30,56,111,131,135],"outcome":[31,47],"reward":[32,76],"(ORMs),":[34],"which":[35],"suffer":[36],"from":[37],"a":[38,116,150],"critical":[39],"granularity":[40],"mismatch:":[41],"ORMs":[42],"are":[43],"designed":[44],"to":[45,59,64],"provide":[46],"rewards":[48,58],"complete":[50,99],"responses,":[51,100],"while":[52],"RGS":[53,80,147],"process":[57,75],"guide":[60],"the":[61],"policy,":[62],"leading":[63],"inconsistent":[65],"scoring":[66],"suboptimal":[68],"alignment.To":[69],"address":[70],"this":[71,163],"challenge,":[72],"we":[73,113],"introduce":[74],"(PRMs)":[78],"into":[79],"argue":[82],"that":[83,142],"an":[84],"ideal":[85],"PRM":[86],"should":[87],"satisfy":[88],"two":[89],"objectives:":[90],"Score":[91],"Consistency,":[92,103],"ensuring":[93],"coherent":[94],"evaluation":[95,127,154],"across":[96,156],"partial":[97,105,126],"Preference":[102],"sequence":[106],"assessments":[107],"preferences.Based":[110],"these,":[112],"propose":[114],"SP-PRM,":[115],"novel":[117],"dual-consistency":[118],"framework":[119],"integrating":[120],"score":[121],"consistency-based":[122,125],"preference":[124],"modules":[128],"without":[129],"relying":[130],"annotation.Extensive":[133],"experiments":[134],"dialogue,":[136],"summarization,":[137],"reasoning":[139],"tasks":[140],"demonstrate":[141],"SP-PRM":[143],"substantially":[144],"enhances":[145],"methods,":[148],"achieving":[149],"3.6%-10.3%improvement":[151],"GPT-4":[153],"scores":[155],"all":[157],"tasks.Code":[158],"is":[159],"publicly":[160],"available":[161],"at":[162],"link.":[164]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}