{"id":"https://openalex.org/W7166841071","doi":"https://doi.org/10.18653/v1/2026.acl-long.1744","title":"Efficient Process Reward Modeling via Contrastive Mutual Information","display_name":"Efficient Process Reward Modeling via Contrastive Mutual Information","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166841071","doi":"https://doi.org/10.18653/v1/2026.acl-long.1744"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.1744","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1744","pdf_url":"https://aclanthology.org/2026.acl-long.1744.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 64th 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/2026.acl-long.1744.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139778615","display_name":"Nakyung Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Nakyung Lee","raw_affiliation_strings":["Department of Electrical and Computer Engineering , Seoul National University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering , Seoul National University ,","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139836045","display_name":"Sangwoo Hong","orcid":null},"institutions":[{"id":"https://openalex.org/I24062138","display_name":"Konkuk University","ror":"https://ror.org/025h1m602","country_code":"KR","type":"education","lineage":["https://openalex.org/I24062138"]},{"id":"https://openalex.org/I4210091929","display_name":"Konkuk University Medical Center","ror":"https://ror.org/00jcx1769","country_code":"KR","type":"healthcare","lineage":["https://openalex.org/I24062138","https://openalex.org/I4210091929"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sangwoo Hong","raw_affiliation_strings":["Department of Computer Science and Engineering , Konkuk University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering , Konkuk University ,","institution_ids":["https://openalex.org/I24062138","https://openalex.org/I4210091929"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139739922","display_name":"Jungwoo Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jungwoo Lee","raw_affiliation_strings":["Department of Electrical and Computer Engineering , Seoul National University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering , Seoul National University ,","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"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":0.84191176,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"37603","last_page":"37617"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10703","display_name":"Business Process Modeling and Analysis","score":0.7876999974250793,"subfield":{"id":"https://openalex.org/subfields/1404","display_name":"Management Information Systems"},"field":{"id":"https://openalex.org/fields/14","display_name":"Business, Management and Accounting"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10703","display_name":"Business Process Modeling and Analysis","score":0.7876999974250793,"subfield":{"id":"https://openalex.org/subfields/1404","display_name":"Management Information Systems"},"field":{"id":"https://openalex.org/fields/14","display_name":"Business, Management and Accounting"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11719","display_name":"Data Quality and Management","score":0.0575999990105629,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.008799999952316284,"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/mutual-information","display_name":"Mutual information","score":0.6238999962806702},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.6008999943733215},{"id":"https://openalex.org/keywords/pointwise-mutual-information","display_name":"Pointwise mutual information","score":0.34360000491142273},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.3124000132083893},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.29190000891685486},{"id":"https://openalex.org/keywords/complete-information","display_name":"Complete information","score":0.2818000018596649},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.2793000042438507}],"concepts":[{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.6238999962806702},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6119999885559082},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.6008999943733215},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5254999995231628},{"id":"https://openalex.org/C7797323","wikidata":"https://www.wikidata.org/wiki/Q3798612","display_name":"Pointwise mutual information","level":3,"score":0.34360000491142273},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3124000132083893},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29190000891685486},{"id":"https://openalex.org/C113336015","wikidata":"https://www.wikidata.org/wiki/Q574010","display_name":"Complete information","level":2,"score":0.2818000018596649},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2793000042438507},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.2703000009059906},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.2676999866962433},{"id":"https://openalex.org/C180198813","wikidata":"https://www.wikidata.org/wiki/Q121182","display_name":"Information system","level":2,"score":0.2615000009536743},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2597000002861023},{"id":"https://openalex.org/C21338462","wikidata":"https://www.wikidata.org/wiki/Q1662581","display_name":"Information model","level":2,"score":0.2563000023365021},{"id":"https://openalex.org/C87868495","wikidata":"https://www.wikidata.org/wiki/Q750843","display_name":"Information processing","level":2,"score":0.25529998540878296},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.25270000100135803}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.1744","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1744","pdf_url":"https://aclanthology.org/2026.acl-long.1744.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.1744","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1744","pdf_url":"https://aclanthology.org/2026.acl-long.1744.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G205224102","display_name":null,"funder_award_id":"RS-2024-00451435","funder_id":"https://openalex.org/F4320322120","funder_display_name":"National Research Foundation of Korea"},{"id":"https://openalex.org/G2920814948","display_name":null,"funder_award_id":"BK21 FOUR","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G2947324332","display_name":null,"funder_award_id":"RS-2021II211343","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G30685149","display_name":null,"funder_award_id":"BK21 FOUR","funder_id":"https://openalex.org/F4320320671","funder_display_name":"National Research Foundation"},{"id":"https://openalex.org/G4475359334","display_name":null,"funder_award_id":"RS2021-II211343","funder_id":"https://openalex.org/F4320321292","funder_display_name":"Seoul National University"},{"id":"https://openalex.org/G45824782","display_name":null,"funder_award_id":"RS-2024-00413957","funder_id":"https://openalex.org/F4320320671","funder_display_name":"National Research Foundation"},{"id":"https://openalex.org/G4603601854","display_name":null,"funder_award_id":"RS-2025-02305453","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G6120464043","display_name":null,"funder_award_id":"RS2021-II211343","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G6390571125","display_name":null,"funder_award_id":"RS-2024-00413957","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G754601064","display_name":null,"funder_award_id":"RS\u20102024\u201000413957","funder_id":"https://openalex.org/F4320322120","funder_display_name":"National Research Foundation of Korea"},{"id":"https://openalex.org/G802237840","display_name":null,"funder_award_id":"RS-2025-02273157","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G8663466005","display_name":null,"funder_award_id":"RS-2025-25442149","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G8978267545","display_name":null,"funder_award_id":"RS-2021-II211343","funder_id":"https://openalex.org/F4320322120","funder_display_name":"National Research Foundation of Korea"}],"funders":[{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"},{"id":"https://openalex.org/F4320321292","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542"},{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"},{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166841071.pdf","grobid_xml":"https://content.openalex.org/works/W7166841071.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"research":[1],"has":[2],"devoted":[3],"considerable":[4],"effort":[5],"to":[6,32,36,60,86,118,131,158],"verifying":[7],"the":[8,82,93,106,110,113,128,132],"intermediate":[9],"reasoning":[10,38,103,170],"steps":[11],"of":[12,96],"chain-of-thought":[13],"(CoT)":[14],"trajectories":[15],"using":[16],"process":[17],"reward":[18,34,77],"models":[19],"(PRMs)":[20],"and":[21,44,112,135,152,168],"other":[22],"verifier":[23],"models.However,":[24],"training":[25],"a":[26,74,102,125,137],"PRM":[27],"typically":[28],"requires":[29],"human":[30],"annotators":[31],"assign":[33],"scores":[35],"each":[37],"step,":[39],"which":[40],"is":[41,173],"both":[42],"costly":[43],"time-consuming.Existing":[45],"automated":[46],"approaches,":[47],"such":[48],"as":[49,124],"Monte":[50],"Carlo":[51],"(MC)":[52],"estimation,":[53,160],"also":[54],"demand":[55],"substantial":[56],"computational":[57,94],"resources":[58],"due":[59],"repeated":[61],"LLM":[62],"rollouts.To":[63],"overcome":[64],"these":[65],"limitations,":[66],"we":[67],"propose":[68],"contrastive":[69,121],"pointwise":[70],"mutual":[71,107],"information":[72,108],"(CPMI),":[73],"novel":[75],"automatic":[76],"labeling":[78,145],"method":[79],"that":[80,143],"leverages":[81],"model's":[83],"internal":[84],"probability":[85],"infer":[87],"step-level":[88],"supervision":[89],"while":[90,161],"significantly":[91],"reducing":[92],"burden":[95],"annotating":[97],"dataset.CPMI":[98],"quantifies":[99],"how":[100],"much":[101],"step":[104,111],"increases":[105],"between":[109],"correct":[114],"target":[115],"answer":[116],"relative":[117],"hard-negative":[119],"alternatives.This":[120],"signal":[122],"serves":[123],"proxy":[126],"for":[127],"step's":[129],"contribution":[130],"final":[133],"solution":[134],"yields":[136],"reliable":[138],"reward.The":[139],"experimental":[140],"results":[141],"show":[142],"CPMI-based":[144],"reduces":[146],"dataset":[147],"construction":[148],"time":[149],"by":[150,155],"84%":[151],"token":[153],"generation":[154],"98%":[156],"compared":[157],"MC":[159],"achieving":[162],"higher":[163],"accuracy":[164],"on":[165],"process-level":[166],"evaluations":[167],"mathematical":[169],"benchmarks.The":[171],"code":[172],"available":[174],"at":[175],"https://github.com/nakyungLee20/CPMI.":[176]},"counts_by_year":[],"updated_date":"2026-08-24T07:32:12.397491","created_date":"2026-07-02T00:00:00"}
