{"id":"https://openalex.org/W7140076372","doi":"https://doi.org/10.18653/v1/2026.eacl-long.251","title":"Detecting Non-Membership in LLM Training Data via Rank Correlations","display_name":"Detecting Non-Membership in LLM Training Data via Rank Correlations","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140076372","doi":"https://doi.org/10.18653/v1/2026.eacl-long.251"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.eacl-long.251","doi":"https://doi.org/10.18653/v1/2026.eacl-long.251","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.251","pdf_url":null,"source":{"id":"https://openalex.org/S4306418011","display_name":"Conference of the European Chapter of the Association for Computational Linguistics","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.18653/v1/2026.eacl-long.251","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125951086","display_name":"Pranav Shetty","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pranav Shetty","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016858094","display_name":"Mirazul Haque","orcid":"https://orcid.org/0009-0006-6627-7688","observed_orcids":["https://orcid.org/0009-0006-6627-7688"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mirazul Haque","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5157360767","display_name":"Zhiqiang Ma","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhiqiang Ma","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130326850","display_name":"Xiaomo Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaomo Liu","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.13993524,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"5414","last_page":"5429"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.6812000274658203,"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":0.6812000274658203,"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":0.07159999758005142,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.05640000104904175,"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/membership-inference","display_name":"membership inference","score":0.7620000243186951},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.765999972820282},{"id":"https://openalex.org/keywords/rank-correlation","display_name":"rank correlation","score":0.7829999923706055},{"id":"https://openalex.org/keywords/prism","display_name":"PRISM","score":0.6790000200271606},{"id":"https://openalex.org/keywords/copyright-enforcement","display_name":"copyright enforcement","score":0.7179999947547913},{"id":"https://openalex.org/keywords/compliance-audit","display_name":"compliance audit","score":0.6240000128746033},{"id":"https://openalex.org/keywords/model-auditing","display_name":"model auditing","score":0.31299999356269836}],"concepts":[{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4603999853134155},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.41909998655319214},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3628000020980835},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.33489999175071716},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3294000029563904},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.3158999979496002},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2791000008583069},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2727000117301941},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26829999685287476},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2551000118255615}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.eacl-long.251","doi":"https://doi.org/10.18653/v1/2026.eacl-long.251","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.251","pdf_url":null,"source":{"id":"https://openalex.org/S4306418011","display_name":"Conference of the European Chapter of the Association for Computational Linguistics","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2603.22707","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2603.22707","pdf_url":"https://arxiv.org/pdf/2603.22707","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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2603.22707","doi":"https://doi.org/10.48550/arxiv.2603.22707","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.22707","pdf_url":null,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.eacl-long.251","doi":"https://doi.org/10.18653/v1/2026.eacl-long.251","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.251","pdf_url":null,"source":{"id":"https://openalex.org/S4306418011","display_name":"Conference of the European Chapter of the Association for Computational Linguistics","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415795489","https://openalex.org/W7220383128","https://openalex.org/W6929500085","https://openalex.org/W7155164840","https://openalex.org/W7130736677","https://openalex.org/W4403556201","https://openalex.org/W4410609172","https://openalex.org/W4403579122","https://openalex.org/W4409657193","https://openalex.org/W4404782717"],"abstract_inverted_index":{"As":[0],"large":[1],"language":[2],"models":[3,88],"(LLMs)":[4],"are":[5],"trained":[6,111],"on":[7,35,112],"increasingly":[8],"vast":[9],"and":[10,28],"opaque":[11],"text":[12],"corpora,":[13],"determining":[14],"which":[15],"data":[16,134],"contributed":[17],"to":[18,79],"training":[19,43,133],"has":[20,58,109],"become":[21],"essential":[22],"for":[23,147],"copyright":[24],"enforcement,":[25],"compliance":[26],"auditing,":[27],"user":[29],"trust.":[30],"While":[31],"prior":[32],"work":[33],"focuses":[34],"detecting":[36],"whether":[37],"a":[38,52,69,93,120,145],"dataset":[39,53,94],"was":[40,54],"used":[41,56],"in":[42,99,132],"(membership":[44],"inference),":[45],"the":[46],"complementary":[47],"problem":[48],"--":[49,57],"verifying":[50,148],"that":[51,71,86,89,113,123,149],"not":[55,91],"received":[59],"little":[60],"attention.":[61],"We":[62],"address":[63],"this":[64,116],"gap":[65],"by":[66],"introducing":[67],"PRISM,":[68],"test":[70,122],"detects":[72,124],"dataset-level":[73],"non-membership":[74],"using":[75],"only":[76],"grey-box":[77],"access":[78],"model":[80,108],"logits.":[81],"Our":[82],"key":[83],"insight":[84],"is":[85],"two":[87],"have":[90],"seen":[92],"exhibit":[95],"higher":[96],"rank":[97],"correlation":[98],"their":[100],"normalized":[101],"token":[102],"log":[103],"probabilities":[104],"than":[105],"when":[106],"one":[107],"been":[110],"data.":[114],"Using":[115],"observation,":[117],"we":[118],"construct":[119],"correlation-based":[121],"non-membership.":[125],"Empirically,":[126],"PRISM":[127],"reliably":[128],"rules":[129],"out":[130],"membership":[131],"across":[135],"all":[136],"datasets":[137,151],"tested":[138],"while":[139],"avoiding":[140],"false":[141],"positives,":[142],"thus":[143],"offering":[144],"framework":[146],"specific":[150],"were":[152],"excluded":[153],"from":[154],"LLM":[155],"training.":[156]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-03-24T00:00:00"}