{"id":"https://openalex.org/W7164026880","doi":"https://doi.org/10.48550/arxiv.2606.07874","title":"Safety is Contextual, LLM-Judges Are Not: Navigating the Rigid Priors of Evaluators","display_name":"Safety is Contextual, LLM-Judges Are Not: Navigating the Rigid Priors of Evaluators","publication_year":2026,"publication_date":"2026-06-05","ids":{"openalex":"https://openalex.org/W7164026880","doi":"https://doi.org/10.48550/arxiv.2606.07874"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.07874","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.07874","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2606.07874","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5046528504","display_name":"Anissa Alloula","orcid":"https://orcid.org/0000-0003-1525-3994"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Alloula, Anissa","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138261259","display_name":"Federico Licini","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Licini, Federico","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138287302","display_name":"Ava Batchkala","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Batchkala, Ava","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5099670415","display_name":"Seraphina Goldfarb-Tarrant","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Goldfarb-Tarrant, Seraphina","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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.17790000140666962,"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"}},"topics":[{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.17790000140666962,"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"}},{"id":"https://openalex.org/T13643","display_name":"Artificial Intelligence in Law","score":0.12200000137090683,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12755","display_name":"Legal Education and Practice Innovations","score":0.04170000180602074,"subfield":{"id":"https://openalex.org/subfields/3308","display_name":"Law"},"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/context","display_name":"Context (archaeology)","score":0.6710000038146973},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6467000246047974},{"id":"https://openalex.org/keywords/workplace-safety","display_name":"Workplace safety","score":0.390500009059906},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.35589998960494995},{"id":"https://openalex.org/keywords/accident","display_name":"Accident (philosophy)","score":0.3111000061035156},{"id":"https://openalex.org/keywords/system-safety","display_name":"System safety","score":0.3073999881744385},{"id":"https://openalex.org/keywords/event","display_name":"Event (particle physics)","score":0.2865000069141388}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6710000038146973},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6467000246047974},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.574999988079071},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5285999774932861},{"id":"https://openalex.org/C2994198235","wikidata":"https://www.wikidata.org/wiki/Q14493193","display_name":"Workplace safety","level":3,"score":0.390500009059906},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.35589998960494995},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.33649998903274536},{"id":"https://openalex.org/C2780289543","wikidata":"https://www.wikidata.org/wiki/Q424630","display_name":"Accident (philosophy)","level":2,"score":0.3111000061035156},{"id":"https://openalex.org/C132835097","wikidata":"https://www.wikidata.org/wiki/Q7663745","display_name":"System safety","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.2865000069141388},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.28439998626708984},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.27889999747276306},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2685999870300293},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.2685999870300293},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.26840001344680786},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.26809999346733093},{"id":"https://openalex.org/C12174686","wikidata":"https://www.wikidata.org/wiki/Q1058438","display_name":"Risk assessment","level":2,"score":0.2671000063419342},{"id":"https://openalex.org/C2776654903","wikidata":"https://www.wikidata.org/wiki/Q2601463","display_name":"SAFER","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.25609999895095825},{"id":"https://openalex.org/C3017944768","wikidata":"https://www.wikidata.org/wiki/Q1450463","display_name":"Poison control","level":2,"score":0.2502000033855438}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.07874","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.07874","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.07874","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.07874","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.6464315056800842,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"LLMs-as-judges":[0],"are":[1,15,96],"the":[2,60,73,104],"only":[3],"way":[4],"to":[5,37,45,99],"evaluate":[6,59],"safety":[7,47,56,61,83,107],"at":[8],"scale.":[9],"Despite":[10],"their":[11,35,43,54,101,110],"importance,":[12],"LLM-judges":[13,89],"themselves":[14],"rarely":[16],"evaluated":[17],"beyond":[18],"human":[19],"agreement":[20],"in":[21,40],"simple,":[22],"static":[23],"benchmarks.":[24],"We":[25,58,85],"therefore":[26],"investigate":[27,72],"two":[28],"under-explored":[29],"but":[30],"crucial":[31],"properties":[32],"of":[33,64,75],"LLMs-as-judges:":[34],"susceptibility":[36],"relying":[38],"on":[39],"context-information,":[41],"and":[42,68,71,81],"steerability":[44],"differing":[46],"definitions,":[48],"which":[49],"may":[50],"not":[51],"align":[52],"with":[53],"internal":[55],"priors.":[57],"judging":[62],"abilities":[63],"many":[65],"generalist":[66],"LLMs":[67],"safety-specific":[69],"judges,":[70],"impact":[74],"task":[76],"demonstrations,":[77],"novel":[78],"in-context":[79],"information,":[80,94],"changing":[82],"definitions.":[84],"find":[86],"that":[87],"while":[88],"can":[90],"learn":[91],"from":[92],"new":[93],"they":[95],"broadly":[97],"unlikely":[98],"adjust":[100],"evaluations":[102],"if":[103],"context":[105],"or":[106],"definition":[108],"contradicts":[109],"prior.":[111]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-10T00:00:00"}