{"id":"https://openalex.org/W7169845001","doi":"https://doi.org/10.3390/computers15070460","title":"SafeBoundary-LLM: Measuring Safety Boundary Stability in Local Open-Weight LLMs Through Single-Turn Baselines and Multi-Turn Escalation","display_name":"SafeBoundary-LLM: Measuring Safety Boundary Stability in Local Open-Weight LLMs Through Single-Turn Baselines and Multi-Turn Escalation","publication_year":2026,"publication_date":"2026-07-21","ids":{"openalex":"https://openalex.org/W7169845001","doi":"https://doi.org/10.3390/computers15070460"},"language":"en","primary_location":{"id":"doi:10.3390/computers15070460","is_oa":true,"landing_page_url":"https://doi.org/10.3390/computers15070460","pdf_url":"https://www.mdpi.com/2073-431X/15/7/460/pdf?version=1784614532","source":{"id":"https://openalex.org/S4210228075","display_name":"Computers","issn_l":"2073-431X","issn":["2073-431X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310310987","host_organization_name":"Multidisciplinary Digital Publishing Institute","host_organization_lineage":["https://openalex.org/P4310310987"],"host_organization_lineage_names":["Multidisciplinary Digital Publishing Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computers","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.mdpi.com/2073-431X/15/7/460/pdf?version=1784614532","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119223040","display_name":"Andreea Alexandra Anghel","orcid":"https://orcid.org/0009-0008-2537-6713"},"institutions":[{"id":"https://openalex.org/I3122748864","display_name":"\"Dunarea de Jos\" University of Galati","ror":"https://ror.org/052sta926","country_code":"RO","type":"education","lineage":["https://openalex.org/I3122748864"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Andreea Alexandra Anghel","raw_affiliation_strings":["Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania"],"raw_orcid":"https://orcid.org/0009-0008-2537-6713","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania","institution_ids":["https://openalex.org/I3122748864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049418389","display_name":"C\u0103t\u0103lin Anghel","orcid":"https://orcid.org/0000-0002-1849-3072"},"institutions":[{"id":"https://openalex.org/I3122748864","display_name":"\"Dunarea de Jos\" University of Galati","ror":"https://ror.org/052sta926","country_code":"RO","type":"education","lineage":["https://openalex.org/I3122748864"]}],"countries":["RO"],"is_corresponding":true,"raw_author_name":"Catalin Anghel","raw_affiliation_strings":["Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania"],"raw_orcid":"https://orcid.org/0000-0002-1849-3072","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania","institution_ids":["https://openalex.org/I3122748864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052924860","display_name":"Emilia Pecheanu","orcid":"https://orcid.org/0000-0003-1155-5274"},"institutions":[{"id":"https://openalex.org/I3122748864","display_name":"\"Dunarea de Jos\" University of Galati","ror":"https://ror.org/052sta926","country_code":"RO","type":"education","lineage":["https://openalex.org/I3122748864"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Emilia Pecheanu","raw_affiliation_strings":["Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania"],"raw_orcid":"https://orcid.org/0000-0003-1155-5274","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania","institution_ids":["https://openalex.org/I3122748864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120336926","display_name":"Antonio Stefan Balau","orcid":"https://orcid.org/0009-0002-5261-2060"},"institutions":[{"id":"https://openalex.org/I3122748864","display_name":"\"Dunarea de Jos\" University of Galati","ror":"https://ror.org/052sta926","country_code":"RO","type":"education","lineage":["https://openalex.org/I3122748864"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Antonio Stefan Balau","raw_affiliation_strings":["Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania"],"raw_orcid":"https://orcid.org/0009-0002-5261-2060","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania","institution_ids":["https://openalex.org/I3122748864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110800740","display_name":"Marian Cr\u0103ciun","orcid":"https://orcid.org/0009-0006-9970-2556"},"institutions":[{"id":"https://openalex.org/I3122748864","display_name":"\"Dunarea de Jos\" University of Galati","ror":"https://ror.org/052sta926","country_code":"RO","type":"education","lineage":["https://openalex.org/I3122748864"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Marian Viorel Craciun","raw_affiliation_strings":["Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania"],"raw_orcid":"https://orcid.org/0009-0006-9970-2556","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania","institution_ids":["https://openalex.org/I3122748864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103039137","display_name":"Adina Cocu","orcid":"https://orcid.org/0000-0003-0935-4713"},"institutions":[{"id":"https://openalex.org/I3122748864","display_name":"\"Dunarea de Jos\" University of Galati","ror":"https://ror.org/052sta926","country_code":"RO","type":"education","lineage":["https://openalex.org/I3122748864"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Adina Cocu","raw_affiliation_strings":["Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania"],"raw_orcid":"https://orcid.org/0000-0003-0935-4713","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania","institution_ids":["https://openalex.org/I3122748864"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5109532282","display_name":"Cristian Sandu","orcid":null},"institutions":[{"id":"https://openalex.org/I3122748864","display_name":"\"Dunarea de Jos\" University of Galati","ror":"https://ror.org/052sta926","country_code":"RO","type":"education","lineage":["https://openalex.org/I3122748864"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Cristian Sandu","raw_affiliation_strings":["Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania"],"raw_orcid":"https://orcid.org/0009-0006-6245-3621","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Technology, \u201cDun\u0103rea de Jos\u201d University of Galati, \u0218tiin\u021bei St. 2, 800146 Galati, Romania","institution_ids":["https://openalex.org/I3122748864"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5049418389"],"corresponding_institution_ids":["https://openalex.org/I3122748864"],"apc_list":{"value":1800,"currency":"CHF","value_usd":2172},"apc_paid":{"value":1800,"currency":"CHF","value_usd":2172},"fwci":7.0482,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.97445394,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":"15","issue":"7","first_page":"460","last_page":"460"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.20239999890327454,"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/T10028","display_name":"Topic Modeling","score":0.20239999890327454,"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/T12262","display_name":"Hate Speech and Cyberbullying Detection","score":0.10429999977350235,"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/T12380","display_name":"Authorship Attribution and Profiling","score":0.07519999891519547,"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/boundary","display_name":"Boundary (topology)","score":0.6425999999046326},{"id":"https://openalex.org/keywords/audit","display_name":"Audit","score":0.6266999840736389},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.6069999933242798},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.5353000164031982},{"id":"https://openalex.org/keywords/compliance","display_name":"Compliance (psychology)","score":0.4196000099182129},{"id":"https://openalex.org/keywords/boundary-line","display_name":"Boundary line","score":0.38199999928474426},{"id":"https://openalex.org/keywords/risk-assessment","display_name":"Risk assessment","score":0.3172999918460846}],"concepts":[{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.6425999999046326},{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.6266999840736389},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.6069999933242798},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.5353000164031982},{"id":"https://openalex.org/C2781460075","wikidata":"https://www.wikidata.org/wiki/Q1399332","display_name":"Compliance (psychology)","level":2,"score":0.4196000099182129},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.41769999265670776},{"id":"https://openalex.org/C2993043157","wikidata":"https://www.wikidata.org/wiki/Q4949967","display_name":"Boundary line","level":2,"score":0.38199999928474426},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.34860000014305115},{"id":"https://openalex.org/C12174686","wikidata":"https://www.wikidata.org/wiki/Q1058438","display_name":"Risk assessment","level":2,"score":0.3172999918460846},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.3172000050544739},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.31470000743865967},{"id":"https://openalex.org/C44249647","wikidata":"https://www.wikidata.org/wiki/Q208498","display_name":"Confidence interval","level":2,"score":0.31220000982284546},{"id":"https://openalex.org/C109986646","wikidata":"https://www.wikidata.org/wiki/Q546113","display_name":"Public policy","level":2,"score":0.3012999892234802},{"id":"https://openalex.org/C162118730","wikidata":"https://www.wikidata.org/wiki/Q1128453","display_name":"Actuarial science","level":1,"score":0.30000001192092896},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.2930000126361847},{"id":"https://openalex.org/C2780233690","wikidata":"https://www.wikidata.org/wiki/Q535347","display_name":"Transparency (behavior)","level":2,"score":0.2879999876022339},{"id":"https://openalex.org/C2779240384","wikidata":"https://www.wikidata.org/wiki/Q1561274","display_name":"Safety culture","level":2,"score":0.28299999237060547},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.27970001101493835},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.2694000005722046},{"id":"https://openalex.org/C107826830","wikidata":"https://www.wikidata.org/wiki/Q929380","display_name":"Environmental resource management","level":1,"score":0.26829999685287476},{"id":"https://openalex.org/C39432304","wikidata":"https://www.wikidata.org/wiki/Q188847","display_name":"Environmental science","level":0,"score":0.2676999866962433},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.26170000433921814},{"id":"https://openalex.org/C27548731","wikidata":"https://www.wikidata.org/wiki/Q88272","display_name":"Investment (military)","level":3,"score":0.25929999351501465},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.25920000672340393},{"id":"https://openalex.org/C2779328685","wikidata":"https://www.wikidata.org/wiki/Q1475557","display_name":"Patient safety","level":3,"score":0.2581999897956848},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.2542000114917755},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2533000111579895}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.3390/computers15070460","is_oa":true,"landing_page_url":"https://doi.org/10.3390/computers15070460","pdf_url":"https://www.mdpi.com/2073-431X/15/7/460/pdf?version=1784614532","source":{"id":"https://openalex.org/S4210228075","display_name":"Computers","issn_l":"2073-431X","issn":["2073-431X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310310987","host_organization_name":"Multidisciplinary Digital Publishing Institute","host_organization_lineage":["https://openalex.org/P4310310987"],"host_organization_lineage_names":["Multidisciplinary Digital Publishing Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computers","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:645755db7c2a4697b96eb7cb747bcd41","is_oa":false,"landing_page_url":"https://doaj.org/article/645755db7c2a4697b96eb7cb747bcd41","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Computers, Vol 15, Iss 7, p 460 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.3390/computers15070460","is_oa":true,"landing_page_url":"https://doi.org/10.3390/computers15070460","pdf_url":"https://www.mdpi.com/2073-431X/15/7/460/pdf?version=1784614532","source":{"id":"https://openalex.org/S4210228075","display_name":"Computers","issn_l":"2073-431X","issn":["2073-431X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310310987","host_organization_name":"Multidisciplinary Digital Publishing Institute","host_organization_lineage":["https://openalex.org/P4310310987"],"host_organization_lineage_names":["Multidisciplinary Digital Publishing Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computers","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.7852027416229248,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7169845001.pdf","grobid_xml":"https://content.openalex.org/works/W7169845001.grobid-xml"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W4391136507","https://openalex.org/W4399750638","https://openalex.org/W4401042900","https://openalex.org/W4406205485","https://openalex.org/W4407207546","https://openalex.org/W4412889939","https://openalex.org/W4412995610","https://openalex.org/W4413081720","https://openalex.org/W4413791999","https://openalex.org/W4414013482","https://openalex.org/W4415271252","https://openalex.org/W4415797606","https://openalex.org/W4416093678","https://openalex.org/W6946301872","https://openalex.org/W7117119982","https://openalex.org/W7126420743","https://openalex.org/W7133134643","https://openalex.org/W7133993877","https://openalex.org/W7134908244","https://openalex.org/W7138848763","https://openalex.org/W7155509515","https://openalex.org/W7161761904","https://openalex.org/W7162751746"],"related_works":[],"abstract_inverted_index":{"Local":[0],"open-weight":[1],"large":[2],"language":[3],"models":[4,29,47],"(LLMs)":[5],"are":[6],"increasingly":[7],"used":[8,67,74],"in":[9,82,144,148],"privacy-sensitive":[10],"settings,":[11],"yet":[12],"isolated":[13],"prompts":[14],"may":[15],"not":[16],"reveal":[17],"whether":[18],"safety":[19],"boundaries":[20],"remain":[21],"stable":[22],"during":[23],"conversation.":[24],"SafeBoundary-LLM":[25],"evaluated":[26,49],"seven":[27],"local":[28],"across":[30],"14":[31],"sensitive":[32],"domains,":[33],"84":[34,41],"boundary":[35,121],"sets,":[36],"672":[37],"single-turn":[38,105,170],"prompts,":[39,106],"and":[40,53,57,71,128,146,158,178],"five-turn":[42],"escalation":[43],"conversations;":[44],"the":[45],"same":[46],"were":[48,93],"separately":[50],"on":[51,153],"XSTest":[52,145],"JBB-Behaviors.":[54],"Evaluators":[55],"R1":[56],"R2":[58,65],"independently":[59],"classified":[60],"all":[61],"12,194":[62],"responses,":[63],"with":[64],"labels":[66,73],"for":[68,75,104,132,182],"primary":[69],"outcomes":[70],"unreconciled":[72],"reliability":[76],"analysis.":[77],"Exact":[78],"agreement":[79],"exceeded":[80],"93%":[81],"each":[83],"dataset.":[84],"In":[85],"SafeBoundary-LLM,":[86],"456":[87,136],"out":[88,134],"of":[89,111,135],"7644":[90],"responses":[91],"(5.97%)":[92],"confirmed-or-mixed":[94],"failures.":[95,137],"The":[96],"multi-turn":[97,173],"failure":[98],"rate":[99,109],"was":[100,142,156],"14.69%":[101],"versus":[102],"0.51%":[103],"yielding":[107],"a":[108],"ratio":[110],"28.80":[112],"(95%":[113],"CI":[114],"[21.20,":[115],"43.80];":[116],"Holm-adjusted":[117],"p":[118],"=":[119],"0.0006);":[120],"collapse":[122],"occurred":[123],"only":[124],"at":[125],"Turns":[126],"4\u20135,":[127],"role-play":[129],"bypass":[130],"accounted":[131],"299":[133],"On":[138],"answer-expected":[139],"items,":[140],"over-refusal":[141],"4.34%":[143],"17.29%":[147],"JBB-Behaviors,":[149],"whereas":[150],"unsafe":[151],"compliance":[152],"refusal-expected":[154],"items":[155],"0.36%":[157],"0.86%,":[159],"respectively.":[160],"These":[161],"findings":[162],"support":[163],"an":[164],"evaluation":[165],"strategy":[166],"that":[167],"includes":[168],"public":[169],"benchmarks,":[171],"controlled":[172],"escalation,":[174],"independent":[175],"human":[176],"review,":[177],"traceable":[179],"audit":[180],"records":[181],"locally":[183],"deployed":[184],"LLMs.":[185]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2026-07-22T00:00:00"}
