{"id":"https://openalex.org/W7154456835","doi":"https://doi.org/10.48550/arxiv.2604.12634","title":"RPRA: Predicting an LLM-Judge for Efficient but Performant Inference","display_name":"RPRA: Predicting an LLM-Judge for Efficient but Performant Inference","publication_year":2026,"publication_date":"2026-04-14","ids":{"openalex":"https://openalex.org/W7154456835","doi":"https://doi.org/10.48550/arxiv.2604.12634"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.12634","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12634","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.12634","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5013716857","display_name":"Dylan R. Ashley","orcid":"https://orcid.org/0000-0001-6148-8802"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Ashley, Dylan R.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133703620","display_name":"Ga\u00ebl Le Lan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lan, Ga\u00ebl Le","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133647618","display_name":"Changsheng Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Changsheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022161993","display_name":"Naina Dhingra","orcid":"https://orcid.org/0000-0001-7546-1213"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dhingra, Naina","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133697327","display_name":"Zhipeng Cai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cai, Zhipeng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133719826","display_name":"Ernie Chang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chang, Ernie","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133696175","display_name":"Mingchen Zhuge","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuge, Mingchen","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133707275","display_name":"Yangyang Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shi, Yangyang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133684772","display_name":"Vikas Chandra","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chandra, Vikas","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128424967","display_name":"J\u00fcrgen Schmidhuber","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Schmidhuber, J\u00fcrgen","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":10,"corresponding_author_ids":["https://openalex.org/A5013716857"],"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":0.43160000443458557,"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":0.43160000443458557,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.13289999961853027,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.08299999684095383,"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/ask-price","display_name":"Ask price","score":0.6362000107765198},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5683000087738037},{"id":"https://openalex.org/keywords/face","display_name":"Face (sociological concept)","score":0.4715999960899353},{"id":"https://openalex.org/keywords/predictive-modelling","display_name":"Predictive modelling","score":0.44290000200271606},{"id":"https://openalex.org/keywords/computational-model","display_name":"Computational model","score":0.39969998598098755}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7671999931335449},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.6362000107765198},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6236000061035156},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.616599977016449},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5683000087738037},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.4715999960899353},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.44290000200271606},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.39969998598098755},{"id":"https://openalex.org/C2779714256","wikidata":"https://www.wikidata.org/wiki/Q25305062","display_name":"Multiple Models","level":2,"score":0.3456000089645386},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.27149999141693115},{"id":"https://openalex.org/C2777115002","wikidata":"https://www.wikidata.org/wiki/Q7168246","display_name":"Performance prediction","level":2,"score":0.2689000070095062}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.12634","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12634","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.12634","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12634","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":"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":{"Large":[0],"language":[1],"models":[2,45,70,87,113,148,162,212],"(LLMs)":[3],"face":[4],"a":[5,57],"fundamental":[6],"trade-off":[7,66],"between":[8],"computational":[9],"efficiency":[10],"(e.g.,":[11],"number":[12],"of":[13,41,55,105,188,199],"parameters)":[14],"and":[15,43,83,108,140,194,203,227],"output":[16],"quality,":[17],"especially":[18],"when":[19,49,75,88,154],"deployed":[20],"on":[21,59],"computationally":[22],"limited":[23],"devices":[24],"such":[25,166],"as":[26],"phones":[27],"or":[28,172],"laptops.":[29],"One":[30],"way":[31,223],"to":[32,71,73,85,117,201,215],"address":[33],"this":[34,65,96,99],"challenge":[35],"is":[36],"by":[37,67],"following":[38],"the":[39,103,185,222],"example":[40],"humans":[42],"have":[44],"ask":[46],"for":[47,224],"help":[48],"they":[50,52,76,78,89,93],"believe":[51,77,92],"are":[53],"incapable":[54],"solving":[56],"problem":[58],"their":[60,126,217],"own;":[61],"we":[62,101],"can":[63,79,163,182,213],"overcome":[64],"allowing":[68],"smaller":[69,161,189],"respond":[72],"queries":[74],"provide":[80],"good":[81],"responses,":[82],"deferring":[84],"larger":[86,147],"do":[90],"not":[91],"can.":[94],"To":[95],"end,":[97],"in":[98],"paper,":[100],"investigate":[102],"viability":[104],"Predict-Answer/Act":[106],"(PA)":[107],"Reason-Predict-Reason-Answer/Act":[109],"(RPRA)":[110],"paradigms":[111],"where":[112],"predict":[114,165,216],"--":[115,119],"prior":[116],"responding":[118],"how":[120],"an":[121,136,175],"LLM":[122,157],"judge":[123],"would":[124],"score":[125],"output.":[127],"We":[128],"evaluate":[129],"three":[130],"approaches:":[131],"zero-shot":[132],"prediction,":[133],"prediction":[134,186],"using":[135],"in-context":[137,176],"report":[138,177,192],"card,":[139],"supervised":[141],"fine-tuning.":[142],"Our":[143],"results":[144],"show":[145],"that":[146,211],"(particularly":[149],"reasoning":[150],"models)":[151],"perform":[152],"well":[153,168],"predicting":[155],"generic":[156],"judges":[158,167],"zero-shot,":[159],"while":[160],"reliably":[164],"after":[169],"being":[170],"fine-tuned":[171],"provided":[173],"with":[174,191],"card.":[178],"Altogether,":[179],"both":[180],"approaches":[181],"substantially":[183],"improve":[184],"accuracy":[187],"models,":[190],"cards":[193],"fine-tuning":[195],"achieving":[196],"mean":[197],"improvements":[198],"up":[200],"55%":[202],"52%":[204],"across":[205],"datasets,":[206],"respectively.":[207],"These":[208],"findings":[209],"suggest":[210],"learn":[214],"own":[218],"performance":[219],"limitations,":[220],"paving":[221],"more":[225],"efficient":[226],"self-aware":[228],"AI":[229],"systems.":[230]},"counts_by_year":[],"updated_date":"2026-04-16T06:09:31.884825","created_date":"2026-04-16T00:00:00"}