{"id":"https://openalex.org/W7222852335","doi":null,"title":"Stronger Judges Do Not See More: Deployment-Integrity Blindness in LLM Evaluation","display_name":"Stronger Judges Do Not See More: Deployment-Integrity Blindness in LLM Evaluation","publication_year":2026,"publication_date":"2026-10-15","ids":{"openalex":"https://openalex.org/W7222852335"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.21135","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.21135","pdf_url":"https://arxiv.org/pdf/2610.21135","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.21135","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124850348","display_name":"Jianchang Su","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Su, Jianchang","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":["https://openalex.org/I140172145"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031626457","display_name":"Dawei Xiang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiang, Dawei","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":["https://openalex.org/I140172145"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5158977645","display_name":"Yifan Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhang, Yifan","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":["https://openalex.org/I140172145"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5158971950","display_name":"Wei Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhang, Wei","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":["https://openalex.org/I140172145"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I140172145"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9672999978065491,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9672999978065491,"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/T11424","display_name":"Security and Verification in Computing","score":0.01549999974668026,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10743","display_name":"Software Testing and Debugging Techniques","score":0.0020000000949949026,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/llm-evaluation","display_name":"LLM evaluation","score":0.6320000290870667},{"id":"https://openalex.org/keywords/behavioral-fingerprinting","display_name":"behavioral fingerprinting","score":0.5920000076293945},{"id":"https://openalex.org/keywords/pac-bounds","display_name":"PAC bounds","score":0.47600001096725464},{"id":"https://openalex.org/keywords/llm-based-agents","display_name":"LLM agents","score":0.46700000762939453}],"concepts":[{"id":"https://openalex.org/C2780929884","wikidata":"https://www.wikidata.org/wiki/Q737460","display_name":"Blindness","level":2,"score":0.7261999845504761},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6262999773025513},{"id":"https://openalex.org/C2778355321","wikidata":"https://www.wikidata.org/wiki/Q17079427","display_name":"Identity (music)","level":2,"score":0.5307999849319458},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.4652000069618225},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.46209999918937683},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4514999985694885},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42080000042915344},{"id":"https://openalex.org/C2780813799","wikidata":"https://www.wikidata.org/wiki/Q3274237","display_name":"Zero (linguistics)","level":2,"score":0.4011000096797943},{"id":"https://openalex.org/C9395851","wikidata":"https://www.wikidata.org/wiki/Q177929","display_name":"Stack (abstract data type)","level":2,"score":0.40049999952316284},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.39899998903274536},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.37929999828338623},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.3716999888420105},{"id":"https://openalex.org/C184297639","wikidata":"https://www.wikidata.org/wiki/Q177765","display_name":"Biometrics","level":2,"score":0.3666999936103821},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.34049999713897705},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.3287999927997589},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.30809998512268066},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2987000048160553},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2768000066280365},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.2720000147819519}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:arXiv.org:2610.21135","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.21135","pdf_url":"https://arxiv.org/pdf/2610.21135","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.21135","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.21135","pdf_url":"https://arxiv.org/pdf/2610.21135","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7222852335.pdf","grobid_xml":"https://content.openalex.org/works/W7222852335.grobid-xml"},"referenced_works_count":1,"referenced_works":["https://openalex.org/W4417172194"],"related_works":["https://openalex.org/W4415314106","https://openalex.org/W7202166437","https://openalex.org/W4415280678","https://openalex.org/W7094947999","https://openalex.org/W7165400976","https://openalex.org/W7171494289","https://openalex.org/W7164882581","https://openalex.org/W7220419810","https://openalex.org/W7207546818","https://openalex.org/W4416100856"],"abstract_inverted_index":{"LLM-agent":[0],"evaluation":[1,77,138],"reports":[2],"behavioral":[3],"scores":[4],"but":[5,123],"leaves":[6],"the":[7,10,76,117],"identity":[8,107],"of":[9,126],"deployed":[11],"runtime":[12],"stack":[13,128],"unreported.":[14],"A":[15,87],"tampered":[16,59],"deployment":[17],"that":[18,31,91],"preserves":[19],"evaluated":[20],"outputs":[21,100],"passes":[22],"every":[23],"output-only":[24],"evaluator.":[25],"We":[26,130],"introduce":[27],"DIB-Bench,":[28],"a":[29,71,105,133],"benchmark":[30],"measures":[32],"this":[33],"deployment-integrity":[34],"blindness":[35],"across":[36],"7,560":[37],"scored":[38],"examples,":[39],"six":[40,127],"tamper":[41,73],"types,":[42],"and":[43,70],"an":[44],"evaluator":[45],"ladder":[46],"from":[47,64,99],"regex":[48],"scorers":[49],"through":[50],"LLM":[51],"judges":[52],"to":[53,67],"GPT-5-nano.":[54],"Stronger":[55],"evaluators":[56],"accept":[57],"more":[58],"outputs:":[60],"mean":[61],"acceptance":[62],"rises":[63],"88.6%":[65],"(rule-based)":[66],"94.1%":[68],"(GPT-5-nano),":[69],"trigger-conditional":[72],"invisible":[74],"on":[75],"suite":[78],"produces":[79],"48-100%":[80],"harmful":[81],"outputs,":[82],"rising":[83],"with":[84,93,111],"model":[85,118],"capability.":[86],"PAC-style":[88],"bound":[89],"shows":[90],"tampers":[92],"zero":[94,112],"output":[95],"shift":[96],"are":[97],"undetectable":[98],"at":[101],"any":[102],"sample":[103],"size;":[104],"manifest-hash":[106],"channel":[108],"detects":[109],"them":[110],"samples.":[113],"Behavioral":[114],"fingerprinting":[115],"identifies":[116],"component":[119],"above":[120],"chance":[121],"(76.7%)":[122],"covers":[124],"one":[125],"components.":[129],"also":[131],"propose":[132],"graded":[134],"deployment-identity":[135],"descriptor":[136],"for":[137],"reports.":[139]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-18T00:00:00"}