{"id":"https://openalex.org/W7206205246","doi":"https://doi.org/10.48550/arxiv.2609.00946","title":"Embedded Conditional Independence Tests for Large Language Model Generated Text with an Application to German Parliament Speeches","display_name":"Embedded Conditional Independence Tests for Large Language Model Generated Text with an Application to German Parliament Speeches","publication_year":2026,"publication_date":"2026-09-01","ids":{"openalex":"https://openalex.org/W7206205246","doi":"https://doi.org/10.48550/arxiv.2609.00946"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.00946","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.00946","pdf_url":"https://arxiv.org/pdf/2609.00946","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.00946","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090459890","display_name":"Marco Simnacher","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I39343248","display_name":"Humboldt-Universit\u00e4t zu Berlin","ror":"https://ror.org/01hcx6992","country_code":"DE","type":"education","lineage":["https://openalex.org/I39343248"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Simnacher, Marco","raw_affiliation_strings":["Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany","institution_ids":["https://openalex.org/I39343248"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056231614","display_name":"Georg Keilbar","orcid":"https://orcid.org/0000-0002-5574-3405","observed_orcids":["https://orcid.org/0000-0002-5574-3405"]},"institutions":[{"id":"https://openalex.org/I39343248","display_name":"Humboldt-Universit\u00e4t zu Berlin","ror":"https://ror.org/01hcx6992","country_code":"DE","type":"education","lineage":["https://openalex.org/I39343248"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Keilbar, Georg","raw_affiliation_strings":["Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany","institution_ids":["https://openalex.org/I39343248"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5149138048","display_name":"Benjamin K\u00f6nig","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I39343248","display_name":"Humboldt-Universit\u00e4t zu Berlin","ror":"https://ror.org/01hcx6992","country_code":"DE","type":"education","lineage":["https://openalex.org/I39343248"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"K\u00f6nig, Benjamin","raw_affiliation_strings":["Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany","institution_ids":["https://openalex.org/I39343248"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074204533","display_name":"Christoph Lippert","orcid":"https://orcid.org/0000-0001-6363-2556","observed_orcids":["https://orcid.org/0000-0001-6363-2556"]},"institutions":[{"id":"https://openalex.org/I143288331","display_name":"Hasso Plattner Institute","ror":"https://ror.org/058rn5r42","country_code":"DE","type":"facility","lineage":["https://openalex.org/I143288331","https://openalex.org/I176453806"]},{"id":"https://openalex.org/I39343248","display_name":"Humboldt-Universit\u00e4t zu Berlin","ror":"https://ror.org/01hcx6992","country_code":"DE","type":"education","lineage":["https://openalex.org/I39343248"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Lippert, Christoph","raw_affiliation_strings":["Chair Digital Health & Machine Learning Hasso-Plattner-Institut for Digital Engineering Prof.-Dr.-Helmert-Stra\u00dfe 2-3 , 14482 Potsdam , Germany","Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chair Digital Health & Machine Learning Hasso-Plattner-Institut for Digital Engineering Prof.-Dr.-Helmert-Stra\u00dfe 2-3 , 14482 Potsdam , Germany","institution_ids":["https://openalex.org/I143288331"]},{"raw_affiliation_string":"Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany","institution_ids":["https://openalex.org/I39343248"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016065135","display_name":"Sonja Greven","orcid":"https://orcid.org/0000-0003-0495-850X","observed_orcids":["https://orcid.org/0000-0003-0495-850X"]},"institutions":[{"id":"https://openalex.org/I39343248","display_name":"Humboldt-Universit\u00e4t zu Berlin","ror":"https://ror.org/01hcx6992","country_code":"DE","type":"education","lineage":["https://openalex.org/I39343248"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Greven, Sonja","raw_affiliation_strings":["Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chair of Statistics Humboldt-Universit\u00e4t zu Berlin Spandauer Str. 1 , 10178 Berlin , Germany","institution_ids":["https://openalex.org/I39343248"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":2,"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/T11303","display_name":"Bayesian Modeling and Causal Inference","score":0.18629999458789825,"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/T11303","display_name":"Bayesian Modeling and Causal Inference","score":0.18629999458789825,"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/T13910","display_name":"Computational and Text Analysis Methods","score":0.13009999692440033,"subfield":{"id":"https://openalex.org/subfields/3300","display_name":"General Social Sciences"},"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/T10028","display_name":"Topic Modeling","score":0.10239999741315842,"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/conditional-independence-tests","display_name":"conditional independence testing","score":0.9710000157356262},{"id":"https://openalex.org/keywords/conditional-dependence","display_name":"conditional dependence","score":0.5070000290870667},{"id":"https://openalex.org/keywords/type-i-error-control","display_name":"type I error control","score":0.8539999723434448},{"id":"https://openalex.org/keywords/power-analysis","display_name":"power analysis","score":0.5220000147819519},{"id":"https://openalex.org/keywords/llm-generated-summaries","display_name":"LLM-generated summaries","score":0.6470000147819519},{"id":"https://openalex.org/keywords/speaker-gender","display_name":"speaker gender","score":0.7789999842643738}],"concepts":[{"id":"https://openalex.org/C79772020","wikidata":"https://www.wikidata.org/wiki/Q5159264","display_name":"Conditional independence","level":2,"score":0.6743000149726868},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6053000092506409},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5806999802589417},{"id":"https://openalex.org/C154775046","wikidata":"https://www.wikidata.org/wiki/Q188","display_name":"German","level":2,"score":0.5698999762535095},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5619000196456909},{"id":"https://openalex.org/C35651441","wikidata":"https://www.wikidata.org/wiki/Q625303","display_name":"Independence (probability theory)","level":2,"score":0.5304999947547913},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4860999882221222},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4595000147819519},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4399000108242035},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4327000081539154},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.3928000032901764},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3822000026702881},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.366100013256073},{"id":"https://openalex.org/C2781440851","wikidata":"https://www.wikidata.org/wiki/Q35749","display_name":"Parliament","level":3,"score":0.3450999855995178},{"id":"https://openalex.org/C44492722","wikidata":"https://www.wikidata.org/wiki/Q327069","display_name":"Conditional probability","level":2,"score":0.3183000087738037},{"id":"https://openalex.org/C191988596","wikidata":"https://www.wikidata.org/wiki/Q628374","display_name":"Null hypothesis","level":2,"score":0.301800012588501},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.30070000886917114},{"id":"https://openalex.org/C87007009","wikidata":"https://www.wikidata.org/wiki/Q210832","display_name":"Statistical hypothesis testing","level":2,"score":0.29170000553131104},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.288100004196167},{"id":"https://openalex.org/C195818886","wikidata":"https://www.wikidata.org/wiki/Q5421724","display_name":"Expressive power","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C186215838","wikidata":"https://www.wikidata.org/wiki/Q772232","display_name":"Conditional expectation","level":2,"score":0.28529998660087585},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C43555835","wikidata":"https://www.wikidata.org/wiki/Q2300258","display_name":"Conditional probability distribution","level":2,"score":0.2768999934196472},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.27410000562667847},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26660001277923584},{"id":"https://openalex.org/C203763787","wikidata":"https://www.wikidata.org/wiki/Q371029","display_name":"Null (SQL)","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2621999979019165},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2572999894618988},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2508000135421753}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.00946","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.00946","pdf_url":"https://arxiv.org/pdf/2609.00946","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.00946","doi":"https://doi.org/10.48550/arxiv.2609.00946","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.00946","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.00946","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.00946","pdf_url":"https://arxiv.org/pdf/2609.00946","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G5316435935","display_name":"FOR 5363: KIFOR Integration von Deep Learning und Statistik zum Verst\u00e4ndnis strukturierter biomedizinischer Daten","funder_award_id":"459422098","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"}],"funders":[{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7206205246.pdf","grobid_xml":"https://content.openalex.org/works/W7206205246.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W3121085131","https://openalex.org/W4386081102","https://openalex.org/W4398406656","https://openalex.org/W4402856368","https://openalex.org/W4412886839","https://openalex.org/W4416429971","https://openalex.org/W4417252869","https://openalex.org/W7130359175","https://openalex.org/W7163043553"],"related_works":["https://openalex.org/W7221191964","https://openalex.org/W7140656516","https://openalex.org/W7134963394","https://openalex.org/W4386216194","https://openalex.org/W4318823044","https://openalex.org/W7140277494","https://openalex.org/W7204203740","https://openalex.org/W7166796145","https://openalex.org/W4417465483","https://openalex.org/W7171286756"],"abstract_inverted_index":{"Conditional":[0],"independence":[1],"tests":[2,39],"(CITs)":[3],"test":[4,51,167],"for":[5,43,136,142,150,190,218],"conditional":[6,169],"dependence":[7],"between":[8],"two":[9,154,229],"random":[10,18],"objects":[11],"$X$":[12,55,81,125],"and":[13,82,84,93,126,156,185,198,200,237],"$Y$":[14,67,114],"given":[15,191],"a":[16,58,133,174,195],"third":[17],"object":[19],"$Z$.":[20],"Existing":[21],"CITs":[22,77],"have":[23],"limited":[24],"applicability":[25],"to":[26,89,94,128,161,178,203,212],"high-dimensional":[27],"data,":[28],"especially":[29],"multimodal":[30],"data":[31],"like":[32],"text.":[33],"However,":[34],"we":[35,50,74,216],"show":[36,97,157],"that":[37,132,158,225],"such":[38],"are":[40],"of":[41,102,118,152,187,221,228],"interest":[42],"large":[44],"language":[45],"model":[46],"(LLM)":[47],"outputs,":[48],"where":[49],"whether":[52],"an":[53,65,86],"output":[54],"generated":[56,244],"from":[57,124],"source":[59],"text":[60],"$Z$":[61,69,83,103,110,127],"carries":[62,111],"information":[63,109,232],"about":[64,112,233],"attribute":[66],"beyond":[68,239],"itself.":[70],"For":[71],"this":[72],"purpose,":[73],"propose":[75,173],"embedded":[76,138,166],"(eCITs),":[78],"which":[79],"embed":[80],"apply":[85],"existing":[87],"CIT":[88,134],"the":[90,100,108,116,120,137,143,153,165,188,210,226,234,240],"resulting":[91],"representations":[92],"$Y$.":[95],"We":[96,146,172],"that,":[98],"provided":[99],"embedding":[101,192,222],"is":[104,140],"sufficient,":[105],"i.e.":[106],"retains":[107],"either":[113],"or":[115],"representation":[117],"$X$,":[119],"null":[121],"hypothesis":[122,139],"transfers":[123],"their":[129],"representations,":[130],"so":[131],"valid":[135,141],"original":[144],"one.":[145],"further":[147],"give":[148],"conditions":[149],"equivalence":[151],"hypotheses,":[155],"sufficiency":[159,163],"weakens":[160],"mean":[162,170],"when":[164],"targets":[168],"independence.":[171],"semi-synthetic":[175],"simulation":[176],"design":[177],"assess":[179],"type":[180],"I":[181],"error":[182],"(T1E)":[183],"control":[184],"power":[186],"eCITs":[189,211],"maps":[193,223],"on":[194,206],"specific":[196],"dataset":[197],"task,":[199],"use":[201],"it":[202],"evaluate":[204],"them":[205],"our":[207],"application.":[208],"Applying":[209],"German":[213],"Parliament":[214],"speeches,":[215],"find":[217],"all":[219],"combinations":[220],"considered":[224],"summaries":[227],"LLMs":[230],"contain":[231],"speaker's":[235],"faction":[236],"gender":[238],"speech":[241],"they":[242],"were":[243],"from.":[245]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-03T00:00:00"}