{"id":"https://openalex.org/W7155366255","doi":"https://doi.org/10.48448/zesv-d319","title":"ARREST: Adversarial Resilient Regulation Enhancing Safety and Truth in Large Language Models","display_name":"ARREST: Adversarial Resilient Regulation Enhancing Safety and Truth in Large Language Models","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7155366255","doi":"https://doi.org/10.48448/zesv-d319"},"language":"en","primary_location":{"id":"doi:10.48448/zesv-d319","doi":"https://doi.org/10.48448/zesv-d319","is_oa":true,"landing_page_url":"https://doi.org/10.48448/zesv-d319","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","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":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Audiovisual"},"volume":null,"type":"other","is_traditional":false,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48448/zesv-d319","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134424196","display_name":"Association for Computational Linguistics 2026","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Association for Computational Linguistics 2026","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055678857","display_name":"Arkaprabha Basu","orcid":"https://orcid.org/0000-0003-0136-0887","observed_orcids":["https://orcid.org/0000-0003-0136-0887"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Basu, Arkaprabha","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134364656","display_name":"Swagatam Das","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Das, Swagatam","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122462197","display_name":"Sharanya Dasgupta","orcid":"https://orcid.org/0009-0006-6385-243X","observed_orcids":["https://orcid.org/0009-0006-6385-243X"]},"institutions":[{"id":"https://openalex.org/I6498739","display_name":"Indian Statistical Institute","ror":"https://ror.org/00q2w1j53","country_code":"IN","type":"education","lineage":["https://openalex.org/I6498739"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Dasgupta, Sharanya","raw_affiliation_strings":["Indian Statistical Institute"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Indian Statistical Institute","institution_ids":["https://openalex.org/I6498739"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5122708723","display_name":"Sujoy Nath","orcid":"https://orcid.org/0009-0003-9974-8018","observed_orcids":["https://orcid.org/0009-0003-9974-8018"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nath, Sujoy","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9359999895095825,"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":0.9359999895095825,"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":0.02319999970495701,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.020400000736117363,"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/large-language-models","display_name":"large language models","score":0.9259999990463257},{"id":"https://openalex.org/keywords/hallucination-correction","display_name":"hallucination correction","score":0.8130000233650208},{"id":"https://openalex.org/keywords/adversarial-training","display_name":"adversarial training","score":0.8320000171661377},{"id":"https://openalex.org/keywords/rlhf-alignment","display_name":"RLHF alignment","score":0.4269999861717224}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7027000188827515},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5460000038146973},{"id":"https://openalex.org/C51929080","wikidata":"https://www.wikidata.org/wiki/Q2425187","display_name":"Codebase","level":3,"score":0.49720001220703125},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.4611999988555908},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.4487000107765198},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.44859999418258667},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4447999894618988},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.4296000003814697},{"id":"https://openalex.org/C3017944768","wikidata":"https://www.wikidata.org/wiki/Q1450463","display_name":"Poison control","level":2,"score":0.42089998722076416},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.4059999883174896},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.4020000100135803},{"id":"https://openalex.org/C168031717","wikidata":"https://www.wikidata.org/wiki/Q1530280","display_name":"Balance (ability)","level":2,"score":0.3962000012397766},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.39489999413490295},{"id":"https://openalex.org/C26869875","wikidata":"https://www.wikidata.org/wiki/Q2596417","display_name":"Beneficiary","level":2,"score":0.39340001344680786},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.382099986076355},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.3562000095844269},{"id":"https://openalex.org/C166735990","wikidata":"https://www.wikidata.org/wiki/Q1750812","display_name":"Human factors and ergonomics","level":3,"score":0.35089999437332153},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.33090001344680786},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.3278999924659729},{"id":"https://openalex.org/C98184364","wikidata":"https://www.wikidata.org/wiki/Q1780131","display_name":"Argument (complex analysis)","level":2,"score":0.3127000033855438},{"id":"https://openalex.org/C90734943","wikidata":"https://www.wikidata.org/wiki/Q17008777","display_name":"Conceptualization","level":2,"score":0.3021000027656555},{"id":"https://openalex.org/C511192102","wikidata":"https://www.wikidata.org/wiki/Q5156948","display_name":"Comprehension","level":2,"score":0.2939999997615814},{"id":"https://openalex.org/C125174871","wikidata":"https://www.wikidata.org/wiki/Q1683210","display_name":"Hazard and operability study","level":3,"score":0.2922999858856201},{"id":"https://openalex.org/C2779990267","wikidata":"https://www.wikidata.org/wiki/Q1365664","display_name":"Culpability","level":2,"score":0.28760001063346863},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2849000096321106},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.2831999957561493},{"id":"https://openalex.org/C166052673","wikidata":"https://www.wikidata.org/wiki/Q83021","display_name":"Empirical evidence","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.27070000767707825},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2676999866962433},{"id":"https://openalex.org/C39549134","wikidata":"https://www.wikidata.org/wiki/Q133080","display_name":"Public relations","level":1,"score":0.2653000056743622},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2639999985694885},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.25049999356269836},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48448/zesv-d319","doi":"https://doi.org/10.48448/zesv-d319","is_oa":true,"landing_page_url":"https://doi.org/10.48448/zesv-d319","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","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":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Audiovisual"}],"best_oa_location":{"id":"doi:10.48448/zesv-d319","doi":"https://doi.org/10.48448/zesv-d319","is_oa":true,"landing_page_url":"https://doi.org/10.48448/zesv-d319","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","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":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Audiovisual"},"sustainable_development_goals":[],"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/W4404782179","https://openalex.org/W7140141059","https://openalex.org/W7220466991","https://openalex.org/W6929169979","https://openalex.org/W4417101781","https://openalex.org/W7213252046","https://openalex.org/W7220705723","https://openalex.org/W4417521604","https://openalex.org/W7116322355","https://openalex.org/W6891701211"],"abstract_inverted_index":{"Human":[0],"cognition,":[1],"driven":[2],"by":[3],"complex":[4],"neurochemical":[5],"processes,":[6],"oscillates":[7],"between":[8],"imagination":[9],"and":[10,12,49,59,104,127,134,141],"reality":[11],"learns":[13],"to":[14,21,46,89,99,146,165,173],"self-correct":[15],"whenever":[16],"such":[17],"subtle":[18],"drifts":[19],"lead":[20],"hallucinations":[22],"or":[23],"unsafe":[24,105],"associations.":[25],"In":[26],"recent":[27],"years,":[28],"LLMs":[29,63],"have":[30],"demonstrated":[31],"remarkable":[32],"performance":[33],"in":[34,62,69,96,144],"a":[35,66,129],"wide":[36],"range":[37],"of":[38],"tasks.":[39],"However,":[40],"they":[41],"still":[42],"lack":[43],"human":[44],"cognition":[45],"balance":[47],"factuality":[48],"safety.":[50],"Bearing":[51],"the":[52,91,97,112,117,166],"resemblance,":[53],"we":[54,119],"argue":[55],"that":[56,84,132,153],"both":[57,139],"factual":[58,147],"safety":[60],"failures":[61],"arise":[64],"from":[65],"\"concept":[67],"misalignment''":[68],"their":[70],"internal":[71],"representations":[72],"rather":[73],"than":[74],"addressing":[75],"those":[76],"as":[77],"entirely":[78],"separate":[79],"alignment":[80],"issues.":[81],"We":[82,176],"hypothesize":[83],"an":[85],"external":[86],"network,":[87],"trained":[88],"understand":[90],"fluctuations,":[92],"can":[93],"selectively":[94],"intervene":[95],"model":[98,113],"regulate":[100],"falsehood":[101],"into":[102,107],"truthfulness":[103],"output":[106,109],"safe":[108],"without":[110],"fine-tuning":[111],"parameters":[114],"themselves.":[115],"Reflecting":[116],"hypothesis,":[118],"propose":[120],"ARREST":[121,154],"(Adversarial":[122],"Resilient":[123],"Regulation":[124],"Enhancing":[125],"Safety":[126],"Truth),":[128],"unified":[130],"framework":[131],"identifies":[133],"corrects":[135],"drifted":[136],"features,":[137],"engaging":[138],"soft":[140,170],"hard":[142],"refusals":[143,171],"addition":[145],"corrections.":[148],"Our":[149],"empirical":[150],"results":[151],"show":[152],"not":[155],"only":[156],"regulates":[157],"misalignment":[158],"but":[159],"is":[160],"also":[161],"more":[162],"versatile":[163],"compared":[164],"RLHF-aligned":[167],"modelsin":[168],"generating":[169],"due":[172],"adversarial":[174],"training.":[175],"make":[177],"our":[178],"codebase":[179],"available":[180],"at":[181],"https://anonymous.4open.science/r/ARREST-LLM":[182],".":[183]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-04-24T00:00:00"}