{"id":"https://openalex.org/W7154286280","doi":"https://doi.org/10.48550/arxiv.2604.10072","title":"Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty","display_name":"Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty","publication_year":2026,"publication_date":"2026-04-11","ids":{"openalex":"https://openalex.org/W7154286280","doi":"https://doi.org/10.48550/arxiv.2604.10072"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.10072","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.10072","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.10072","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133598525","display_name":"Chao Xue","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xue, Chao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133584615","display_name":"Yao Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067658211","display_name":"Mengqiao Liu","orcid":"https://orcid.org/0000-0001-8426-3091"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Mengqiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133561897","display_name":"Di Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Di","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133587881","display_name":"Xingsheng Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Xingsheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133572360","display_name":"Peiyang Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Peiyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133569838","display_name":"Xianjie Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xianjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133585249","display_name":"Chenyao Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Chenyao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133561850","display_name":"Lei Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Lei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133600184","display_name":"Yu Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133562699","display_name":"Haibo Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shi, Haibo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133604915","display_name":"Shuang Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Shuang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133622260","display_name":"Minlong Peng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Minlong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133608705","display_name":"Flora D. Salim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Salim, Flora D.","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":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.13779999315738678,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.13779999315738678,"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/T10028","display_name":"Topic Modeling","score":0.13529999554157257,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.10339999943971634,"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/inference","display_name":"Inference","score":0.7127000093460083},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.590499997138977},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.564300000667572},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.4717000126838684},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.4196000099182129},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.40380001068115234}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7591000199317932},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7127000093460083},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6288999915122986},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5946999788284302},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.590499997138977},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.564300000667572},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.4717000126838684},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.4196000099182129},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.40380001068115234},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.39419999718666077},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.37139999866485596},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.31779998540878296},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.30309998989105225}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.10072","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.10072","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.10072","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.10072","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"score":0.7375971674919128,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,79,96],"the":[3,14,101],"Generative":[4],"Reward":[5],"Model":[6],"(GRM)":[7],"have":[8],"demonstrated":[9],"its":[10],"potential":[11],"to":[12,41,57,69,108,143],"enhance":[13],"reasoning":[15,81,115,148,153],"abilities":[16],"of":[17,28,45,104,147],"LLMs":[18],"through":[19],"Chain-of-Thought":[20],"(CoT)":[21],"prompting.":[22],"Despite":[23],"these":[24],"gains,":[25],"existing":[26,62],"implementations":[27],"GRM":[29],"suffer":[30],"from":[31],"two":[32],"critical":[33],"limitations.":[34],"First,":[35],"CoT":[36,71,114],"prompting":[37],"is":[38,171],"applied":[39],"indiscriminately":[40],"all":[42],"inputs":[43],"regardless":[44],"their":[46],"inherent":[47],"complexity.":[48],"This":[49],"introduces":[50],"unnecessary":[51],"computational":[52],"costs":[53],"for":[54,177],"tasks":[55],"amenable":[56],"fast,":[58],"direct":[59],"inference.":[60],"Second,":[61],"approaches":[63],"primarily":[64],"rely":[65],"on":[66,121,151],"voting-based":[67],"mechanisms":[68],"evaluate":[70],"outputs,":[72],"which":[73],"often":[74],"lack":[75],"granularity":[76],"and":[77,111,174],"precision":[78],"assessing":[80],"quality.":[82],"In":[83],"this":[84],"paper,":[85],"we":[86,131],"propose":[87],"E-GRM,":[88],"an":[89,172],"efficient":[90,178],"generative":[91],"reward":[92,129,180],"modeling":[93],"framework":[94],"grounded":[95],"model-internal":[97,169],"uncertainty.":[98],"E-GRM":[99,157],"leverages":[100],"convergence":[102],"behavior":[103],"parallel":[105],"model":[106],"generations":[107],"estimate":[109],"uncertainty":[110,170],"selectively":[112],"trigger":[113],"only":[116],"when":[117],"needed,":[118],"without":[119],"relying":[120],"handcrafted":[122],"features":[123],"or":[124],"task-dependent":[125],"signals.":[126],"To":[127],"improve":[128],"fidelity,":[130],"introduce":[132],"a":[133,139],"lightweight":[134],"discriminative":[135],"scorer":[136],"trained":[137],"with":[138],"hybrid":[140],"regression--ranking":[141],"objective":[142],"provide":[144],"fine-grained":[145],"evaluation":[146],"paths.":[149],"Experiments":[150],"multiple":[152],"benchmarks":[154],"show":[155],"that":[156,168],"substantially":[158],"reduces":[159],"inference":[160],"cost":[161],"while":[162],"consistently":[163],"improving":[164],"answer":[165],"accuracy,":[166],"demonstrating":[167],"effective":[173],"general":[175],"signal":[176],"reasoning-aware":[179],"modeling.":[181]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-15T00:00:00"}