{"id":"https://openalex.org/W7140133907","doi":"https://doi.org/10.18653/v1/2026.eacl-long.212","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":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140133907","doi":"https://doi.org/10.18653/v1/2026.eacl-long.212"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.eacl-long.212","doi":"https://doi.org/10.18653/v1/2026.eacl-long.212","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.212","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","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.18653/v1/2026.eacl-long.212","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","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":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sharanya Dasgupta","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":"Arkaprabha Basu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","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":"Sujoy Nath","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5032906052","display_name":"Shouman Das","orcid":"https://orcid.org/0000-0001-6843-4508","observed_orcids":["https://orcid.org/0000-0001-6843-4508"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Swagatam Das","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":0.10186949,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4565","last_page":"4584"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8765000104904175,"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":0.8765000104904175,"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":0.06390000134706497,"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":0.026200000196695328,"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":0.8980000019073486},{"id":"https://openalex.org/keywords/hallucination-correction","display_name":"hallucination correction","score":0.8149999976158142},{"id":"https://openalex.org/keywords/safety-alignment","display_name":"safety alignment","score":0.7850000262260437},{"id":"https://openalex.org/keywords/adversarial-training","display_name":"adversarial training","score":0.8960000276565552}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.6342999935150146},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5598999857902527},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.48100000619888306},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.40209999680519104},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.35519999265670776},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33070001006126404},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.311599999666214},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.3075000047683716},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.2962999939918518},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.2906999886035919},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.27570000290870667},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.27410000562667847},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.27140000462532043},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.26600000262260437},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2572000026702881},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.25690001249313354},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.25130000710487366}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.eacl-long.212","doi":"https://doi.org/10.18653/v1/2026.eacl-long.212","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.212","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:2601.04394","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2601.04394","pdf_url":"https://arxiv.org/pdf/2601.04394","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"},{"id":"doi:10.48550/arxiv.2601.04394","doi":"https://doi.org/10.48550/arxiv.2601.04394","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.04394","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":"doi:10.18653/v1/2026.eacl-long.212","doi":"https://doi.org/10.18653/v1/2026.eacl-long.212","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.212","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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140133907.pdf","grobid_xml":"https://content.openalex.org/works/W7140133907.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W4384918448","https://openalex.org/W4385374425","https://openalex.org/W4386081218","https://openalex.org/W4386839891","https://openalex.org/W4387355345","https://openalex.org/W4391377022","https://openalex.org/W4391421553","https://openalex.org/W4392617597","https://openalex.org/W4399597480","https://openalex.org/W4405655184","https://openalex.org/W4412158322","https://openalex.org/W7171469682"],"related_works":["https://openalex.org/W7196997594","https://openalex.org/W7147484456","https://openalex.org/W7162426826","https://openalex.org/W7202207464","https://openalex.org/W7154457865","https://openalex.org/W6885142527","https://openalex.org/W7164891000","https://openalex.org/W4415304377","https://openalex.org/W7084128304","https://openalex.org/W7131843780"],"abstract_inverted_index":{"Human":[0],"cognition,":[1],"driven":[2],"by":[3],"complex":[4],"neurochemical":[5],"processes,":[6],"oscillates":[7],"between":[8],"imagination":[9],"and":[10,12,49,59,105,128,135,142],"reality":[11],"learns":[13],"to":[14,21,46,90,100,147,166,175],"self-correct":[15],"whenever":[16],"such":[17],"subtle":[18],"drifts":[19],"lead":[20],"hallucinations":[22],"or":[23],"unsafe":[24,106],"associations.":[25],"In":[26],"recent":[27],"years,":[28],"LLMs":[29,63],"have":[30],"demonstrated":[31],"remarkable":[32],"performance":[33],"in":[34,62,69,97,145,170],"a":[35,66,130],"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":[52,92,98,113,118,167],"resemblance,":[53],"we":[54,120],"argue":[55],"that":[56,85,133,154],"both":[57,140],"factual":[58,148],"safety":[60],"failures":[61],"arise":[64],"from":[65],"representational":[67],"misalignment":[68,159],"their":[70],"latent":[71],"activation":[72],"space,":[73],"rather":[74],"than":[75],"addressing":[76],"those":[77],"as":[78],"entirely":[79],"separate":[80],"alignment":[81],"issues.":[82],"We":[83,178],"hypothesize":[84],"an":[86],"external":[87],"network,":[88],"trained":[89],"understand":[91],"fluctuations,":[93],"can":[94],"selectively":[95],"intervene":[96],"model":[99,114],"regulate":[101],"falsehood":[102],"into":[103,108],"truthfulness":[104],"output":[107,110],"safe":[109],"without":[111],"fine-tuning":[112],"parameters":[115],"themselves.":[116],"Reflecting":[117],"hypothesis,":[119],"propose":[121],"ARREST":[122,155],"(Adversarial":[123],"Resilient":[124],"Regulation":[125],"Enhancing":[126],"Safety":[127],"Truth),":[129],"unified":[131],"framework":[132],"identifies":[134],"corrects":[136],"drifted":[137],"features,":[138],"engaging":[139],"soft":[141,172],"hard":[143],"refusals":[144,173],"addition":[146],"corrections.":[149],"Our":[150],"empirical":[151],"results":[152],"show":[153],"not":[156],"only":[157],"regulates":[158],"but":[160],"is":[161],"also":[162],"more":[163],"versatile":[164],"compared":[165],"RLHF-aligned":[168],"models":[169],"generating":[171],"due":[174],"adversarial":[176],"training.":[177],"make":[179],"our":[180],"codebase":[181],"available":[182],"at":[183],"https://github.com/sharanya-dasgupta001/ARREST.":[184]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-03-24T00:00:00"}