{"id":"https://openalex.org/W7166735687","doi":"https://doi.org/10.48550/arxiv.2606.29784","title":"HERO: Improving the Reliability and Sensitivity of Generative Model Evaluation Using Historical Data","display_name":"HERO: Improving the Reliability and Sensitivity of Generative Model Evaluation Using Historical Data","publication_year":2026,"publication_date":"2026-06-29","ids":{"openalex":"https://openalex.org/W7166735687","doi":"https://doi.org/10.48550/arxiv.2606.29784"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.29784","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.29784","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.2606.29784","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139658635","display_name":"Xinrui Ruan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ruan, Xinrui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107849883","display_name":"Z. Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Zhenyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139636630","display_name":"Waverly Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Waverly","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139712891","display_name":"Yueshan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yueshan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139652560","display_name":"Zeyu Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Zeyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043900893","display_name":"Sicong Huang","orcid":"https://orcid.org/0000-0003-0854-4734"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Sui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5043785463","display_name":"Jingshen Wang","orcid":"https://orcid.org/0000-0002-1432-3834"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jingshen","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/T12535","display_name":"Machine Learning and Data Classification","score":0.4041000008583069,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.4041000008583069,"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/T11704","display_name":"Mobile Crowdsensing and Crowdsourcing","score":0.0794999971985817,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.07559999823570251,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.6917999982833862},{"id":"https://openalex.org/keywords/variance","display_name":"Variance (accounting)","score":0.6484000086784363},{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.5759000182151794},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.531499981880188},{"id":"https://openalex.org/keywords/model-selection","display_name":"Model selection","score":0.3684999942779541},{"id":"https://openalex.org/keywords/hero","display_name":"HERO","score":0.3652999997138977},{"id":"https://openalex.org/keywords/covariate","display_name":"Covariate","score":0.3614000082015991},{"id":"https://openalex.org/keywords/sensitivity","display_name":"Sensitivity (control systems)","score":0.3601999878883362}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7562000155448914},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.6917999982833862},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.6484000086784363},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.5759000182151794},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5514000058174133},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.531499981880188},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5145999789237976},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3961000144481659},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.3684999942779541},{"id":"https://openalex.org/C51364203","wikidata":"https://www.wikidata.org/wiki/Q1563532","display_name":"HERO","level":2,"score":0.3652999997138977},{"id":"https://openalex.org/C119043178","wikidata":"https://www.wikidata.org/wiki/Q320723","display_name":"Covariate","level":2,"score":0.3614000082015991},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.3601999878883362},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.3596999943256378},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.35350000858306885},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.3458999991416931},{"id":"https://openalex.org/C2777338717","wikidata":"https://www.wikidata.org/wiki/Q1762621","display_name":"Vendor","level":2,"score":0.34200000762939453},{"id":"https://openalex.org/C40993552","wikidata":"https://www.wikidata.org/wiki/Q514654","display_name":"Gold standard (test)","level":2,"score":0.3190000057220459},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.31520000100135803},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.3093000054359436},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.29784","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.29784","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.2606.29784","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.29784","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":[{"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth","score":0.45604240894317627}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reliable":[0],"generative":[1],"AI":[2],"models":[3],"critically":[4],"rely":[5],"on":[6,65,232],"expert":[7],"human":[8],"annotations":[9],"to":[10,20,30,74,121,143,175],"evaluate":[11],"output":[12],"quality,":[13],"yet":[14],"these":[15],"\"gold\"":[16],"labels":[17,36,58,69],"are":[18],"expensive":[19],"collect":[21],"and":[22,62,105,148,167,196,218,228],"limited":[23],"in":[24,153,182,207,225],"quantity.":[25],"Organizations":[26],"thus":[27],"often":[28],"turn":[29],"collecting":[31],"vast":[32],"but":[33],"noisy":[34,56],"\"silver\"":[35],"from":[37,163],"crowdsourced":[38],"workers":[39],"or":[40],"vendor":[41],"annotators":[42],"as":[43],"proxies":[44],"for":[45],"gold":[46,49,68,165],"labels.":[47],"Because":[48],"remains":[50,197],"the":[51,76,113,169,183,208,216],"evaluation":[52,85,98,116,194,235],"target,":[53],"naively":[54],"aggregating":[55],"silver":[57,159],"may":[59,70],"introduce":[60,129],"bias,":[61],"estimators":[63],"built":[64],"sparsely":[66],"observed":[67],"have":[71],"high":[72,180],"variance":[73,150,219],"resolve":[75],"model":[77,102,134,154,234],"performance":[78,155,161,224],"gaps":[79],"that":[80,139],"guide":[81],"practical":[82],"decisions.":[83],"Model":[84],"has":[86],"become":[87],"an":[88],"ongoing":[89],"operational":[90],"practice":[91],"rather":[92],"than":[93],"a":[94,136,201],"one-time":[95],"exercise,":[96],"with":[97,179],"rounds":[99],"repeating":[100],"across":[101,191],"versions,":[103],"releases,":[104],"content":[106],"domains.":[107],"A":[108],"natural":[109],"question":[110],"is":[111],"whether":[112],"previous":[114],"historical":[115,141,164,184,204],"data":[117,142],"can":[118,187],"be":[119,188],"used":[120],"improve":[122],"each":[123],"new":[124],"round":[125],"of":[126,203],"evaluation.":[127,156],"We":[128,211],"HERO":[130,157,186],"(History":[131],"Enhanced":[132],"RObust":[133],"evaluation),":[135],"novel":[137],"framework":[138],"uses":[140],"suppress":[144],"bias":[145,217],"(improve":[146,151],"reliability)":[147],"reduce":[149],"sensitivity)":[152],"calibrates":[158],"labelers'":[160],"learned":[162],"annotations,":[166],"stabilizes":[168],"resulting":[170],"estimator":[171],"by":[172],"anchoring":[173],"it":[174],"covariate":[176],"information":[177],"measured":[178],"precision":[181],"data.":[185],"broadly":[189],"applied":[190],"multiple":[192],"common":[193],"tasks,":[195],"valid":[198],"when":[199],"only":[200],"subset":[202],"labelers":[205],"appears":[206],"current":[209],"round.":[210],"establish":[212],"conditions":[213],"under":[214],"which":[215],"reductions":[220],"hold,":[221],"showcase":[222],"HERO's":[223],"simulation":[226],"studies,":[227],"demonstrate":[229],"its":[230],"effectiveness":[231],"real-world":[233],"benchmarking":[236],"datasets.":[237]},"counts_by_year":[],"updated_date":"2026-07-01T06:29:00.853634","created_date":"2026-07-01T00:00:00"}