{"id":"https://openalex.org/W4414894368","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1576","title":"XGUARD: A Graded Benchmark for Evaluating Safety Failures of Large Language Models on Extremist Content","display_name":"XGUARD: A Graded Benchmark for Evaluating Safety Failures of Large Language Models on Extremist Content","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4414894368","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1576"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1576","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1576","pdf_url":"https://aclanthology.org/2026.findings-acl.1576.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1576.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119877696","display_name":"Vadivel Abishethvarman","orcid":null},"institutions":[{"id":"https://openalex.org/I908128300","display_name":"Sabaragamuwa University of Sri Lanka","ror":"https://ror.org/045vwzt11","country_code":"LK","type":"education","lineage":["https://openalex.org/I908128300"]}],"countries":["LK"],"is_corresponding":false,"raw_author_name":"Vadivel Abishethvarman","raw_affiliation_strings":["Sabaragamuwa University of Sri Lanka"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sabaragamuwa University of Sri Lanka","institution_ids":["https://openalex.org/I908128300"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119877697","display_name":"Bhavik Chandna","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bhavik Chandna","raw_affiliation_strings":["University of California , San Diego , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California , San Diego , USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119877698","display_name":"Pratik Jalan","orcid":null},"institutions":[{"id":"https://openalex.org/I99043593","display_name":"Macquarie University","ror":"https://ror.org/01sf06y89","country_code":"AU","type":"education","lineage":["https://openalex.org/I99043593"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Pratik Jalan","raw_affiliation_strings":["Macquarie University , Sydney , Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Macquarie University , Sydney , Australia","institution_ids":["https://openalex.org/I99043593"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077006200","display_name":"Usman Naseem","orcid":"https://orcid.org/0000-0003-0191-7171"},"institutions":[{"id":"https://openalex.org/I99043593","display_name":"Macquarie University","ror":"https://ror.org/01sf06y89","country_code":"AU","type":"education","lineage":["https://openalex.org/I99043593"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Usman Naseem","raw_affiliation_strings":["Macquarie University , Sydney , Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Macquarie University , Sydney , Australia","institution_ids":["https://openalex.org/I99043593"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"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":0.01028673,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"31492","last_page":"31510"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9610000252723694,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9610000252723694,"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/T13629","display_name":"Text Readability and Simplification","score":0.9549999833106995,"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/trustworthiness","display_name":"Trustworthiness","score":0.5756999850273132},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5694000124931335},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5432000160217285},{"id":"https://openalex.org/keywords/ideology","display_name":"Ideology","score":0.46070000529289246},{"id":"https://openalex.org/keywords/rhetoric","display_name":"Rhetoric","score":0.4569999873638153},{"id":"https://openalex.org/keywords/content-analysis","display_name":"Content analysis","score":0.45669999718666077},{"id":"https://openalex.org/keywords/social-media","display_name":"Social media","score":0.3418000042438507},{"id":"https://openalex.org/keywords/threat-model","display_name":"Threat model","score":0.3409999907016754}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6086999773979187},{"id":"https://openalex.org/C153701036","wikidata":"https://www.wikidata.org/wiki/Q659974","display_name":"Trustworthiness","level":2,"score":0.5756999850273132},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5694000124931335},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5432000160217285},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.534600019454956},{"id":"https://openalex.org/C158071213","wikidata":"https://www.wikidata.org/wiki/Q7257","display_name":"Ideology","level":3,"score":0.46070000529289246},{"id":"https://openalex.org/C1370556","wikidata":"https://www.wikidata.org/wiki/Q81009","display_name":"Rhetoric","level":2,"score":0.4569999873638153},{"id":"https://openalex.org/C162446236","wikidata":"https://www.wikidata.org/wiki/Q653137","display_name":"Content analysis","level":2,"score":0.45669999718666077},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.364300012588501},{"id":"https://openalex.org/C518677369","wikidata":"https://www.wikidata.org/wiki/Q202833","display_name":"Social media","level":2,"score":0.3418000042438507},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.3409999907016754},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.30630001425743103},{"id":"https://openalex.org/C3017944768","wikidata":"https://www.wikidata.org/wiki/Q1450463","display_name":"Poison control","level":2,"score":0.2962000072002411},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.28940001130104065},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.28630000352859497},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C65856478","wikidata":"https://www.wikidata.org/wiki/Q3991682","display_name":"Attack model","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C84389358","wikidata":"https://www.wikidata.org/wiki/Q1129466","display_name":"Discourse analysis","level":2,"score":0.25760000944137573},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.25609999895095825}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1576","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1576","pdf_url":"https://aclanthology.org/2026.findings-acl.1576.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2506.00973","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.00973","pdf_url":"https://arxiv.org/pdf/2506.00973","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,"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.2506.00973","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.00973","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1576","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1576","pdf_url":"https://aclanthology.org/2026.findings-acl.1576.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4414894368.pdf","grobid_xml":"https://content.openalex.org/works/W4414894368.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"can":[4],"generate":[5],"content":[6,51],"spanning":[7],"ideological":[8],"rhetoric":[9],"to":[10,45,107],"explicit":[11],"instructions":[12],"for":[13,150],"violence.However,":[14],"existing":[15],"safety":[16,133,148],"evaluations":[17],"often":[18],"rely":[19],"on":[20,55],"simplistic":[21],"binary":[22],"labels":[23],"(safe/unsafe),":[24],"overlooking":[25],"the":[26,47,101,144],"nuanced":[27,93],"spectrum":[28],"of":[29,49,89,95,146],"risk":[30],"these":[31],"outputs":[32],"pose.To":[33],"address":[34],"this,":[35],"we":[36,118],"present":[37],"XGUARD,":[38,117],"a":[39,56],"benchmark":[40],"and":[41,75,98,110,123,135,139,155],"evaluation":[42],"framework":[43,77],"designed":[44],"assess":[46],"severity":[48],"extremist":[50,69,90],"generated":[52,63],"by":[53,67,87],"LLMs":[54,122],"multi-level":[57],"grading.It":[58],"includes":[59],"3,840":[60],"red-teaming":[61],"prompts":[62],"using":[64],"templates":[65],"informed":[66],"real-world":[68],"scenarios":[70],"from":[71],"social":[72],"media,":[73],"forums,":[74],"news.The":[76],"categorizes":[78],"model":[79],"responses":[80],"into":[81,131],"five":[82,120],"danger":[83],"levels":[84],"(0-4)":[85],"defined":[86],"degree":[88],"endorsement,":[91],"enabling":[92],"analysis":[94],"failure":[96],"frequency":[97],"severity.We":[99],"introduce":[100],"interpretable":[102],"Attack":[103],"Severity":[104],"Curve":[105],"(ASC)":[106],"visualize":[108],"vulnerabilities":[109],"compare":[111],"defense":[112,126],"mechanisms":[113],"across":[114],"threat":[115],"intensities.Using":[116],"evaluate":[119],"popular":[121],"two":[124],"lightweight":[125],"strategies,":[127],"revealing":[128],"key":[129],"insights":[130],"current":[132],"gaps":[134],"trade-offs":[136],"between":[137],"robustness":[138],"expressive":[140],"freedom.Our":[141],"work":[142],"underscores":[143],"value":[145],"graded":[147],"metrics":[149],"building":[151],"trustworthy":[152],"LLMs.The":[153],"code":[154],"dataset":[156],"is":[157],"available":[158],"at":[159],"https://github.com/":[160],"Abishethvarman/XGUARD":[161],".":[162]},"counts_by_year":[],"updated_date":"2026-08-18T07:49:30.821534","created_date":"2025-10-10T00:00:00"}