{"id":"https://openalex.org/W7163134644","doi":"https://doi.org/10.48550/arxiv.2606.01166","title":"BraveGuard: From Open-World Threats to Safer Computer-Use Agents","display_name":"BraveGuard: From Open-World Threats to Safer Computer-Use Agents","publication_year":2026,"publication_date":"2026-05-31","ids":{"openalex":"https://openalex.org/W7163134644","doi":"https://doi.org/10.48550/arxiv.2606.01166"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.01166","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01166","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.01166","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103258005","display_name":"Yunhao Feng","orcid":"https://orcid.org/0009-0005-3042-5266"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feng, Yunhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137621211","display_name":"Yifan Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Du, Xiaohu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067405541","display_name":"Xiaohu Du","orcid":"https://orcid.org/0000-0003-4455-3128"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deng, Xinhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137619571","display_name":"Ming Wen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Yifan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137698033","display_name":"Xinhao Deng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wen, Ming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137695432","display_name":"Yanming Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yixu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137624613","display_name":"Yuxiang Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Yuxiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137628413","display_name":"Baihui Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Baihui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137668804","display_name":"Yingshui Tan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan, Yingshui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137617523","display_name":"Yige Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yige","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137675296","display_name":"Yutao Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yutao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137654901","display_name":"Yixu Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Kerui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137657891","display_name":"Kerui Cao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Wenke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137708714","display_name":"Wenke Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Yanming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137647939","display_name":"Xingjun Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Xingjun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137644307","display_name":"Yu-Gang Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Yu-Gang","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":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":0.8223999738693237,"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":0.8223999738693237,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.02800000086426735,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10734","display_name":"Information and Cyber Security","score":0.027799999341368675,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/guard","display_name":"Guard (computer science)","score":0.885699987411499},{"id":"https://openalex.org/keywords/harm","display_name":"Harm","score":0.6604999899864197},{"id":"https://openalex.org/keywords/safer","display_name":"SAFER","score":0.6110000014305115},{"id":"https://openalex.org/keywords/executable","display_name":"Executable","score":0.459199994802475},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.45260000228881836},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.37959998846054077}],"concepts":[{"id":"https://openalex.org/C141141315","wikidata":"https://www.wikidata.org/wiki/Q2379942","display_name":"Guard (computer science)","level":2,"score":0.885699987411499},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.670199990272522},{"id":"https://openalex.org/C2777363581","wikidata":"https://www.wikidata.org/wiki/Q15098235","display_name":"Harm","level":2,"score":0.6604999899864197},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.65829998254776},{"id":"https://openalex.org/C2776654903","wikidata":"https://www.wikidata.org/wiki/Q2601463","display_name":"SAFER","level":2,"score":0.6110000014305115},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.557200014591217},{"id":"https://openalex.org/C160145156","wikidata":"https://www.wikidata.org/wiki/Q778586","display_name":"Executable","level":2,"score":0.459199994802475},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.45260000228881836},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.37959998846054077},{"id":"https://openalex.org/C140547941","wikidata":"https://www.wikidata.org/wiki/Q7797194","display_name":"Threat model","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C2779777834","wikidata":"https://www.wikidata.org/wiki/Q4202277","display_name":"Enforcement","level":2,"score":0.2800999879837036},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.2635999917984009}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.01166","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01166","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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:10.48550/arxiv.2606.01166","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01166","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.6035606265068054,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Computer-use":[0],"agents":[1,210],"extend":[2],"language":[3],"models":[4,59],"from":[5,28,60,165],"text":[6],"generation":[7],"to":[8,26,73,167],"sustained":[9],"interaction":[10],"with":[11,162],"files,":[12],"terminals,":[13],"browsers,":[14],"and":[15,64,77,89,100,131,134,185,196],"external":[16],"tools.":[17],"This":[18],"shift":[19],"creates":[20],"safety":[21,146,191],"risks":[22,76],"that":[23,177],"are":[24],"difficult":[25],"detect":[27],"isolated":[29],"prompts":[30],"or":[31],"final":[32],"responses,":[33],"because":[34],"harm":[35],"often":[36],"emerges":[37],"only":[38],"through":[39],"multi-step":[40],"execution":[41,188],"traces":[42],"whose":[43],"individual":[44],"actions":[45],"appear":[46],"locally":[47],"benign.":[48],"We":[49,121],"introduce":[50],"BraveGuard,":[51],"a":[52,116,202],"self-evolving":[53],"defense":[54,112],"framework":[55],"for":[56,93,208],"training":[57,119,125],"guard":[58,94,127,160,178],"open-world":[61,182],"threat":[62,183],"signals":[63],"realistic":[65,186],"agent":[66,87,187],"trajectories.":[67,150],"BraveGuard":[68,123,143,200],"mines":[69],"recent":[70],"research":[71],"sources":[72],"identify":[74],"emerging":[75],"attack":[78],"patterns,":[79],"instantiates":[80],"them":[81],"as":[82],"executable":[83],"computer-use":[84,149,209],"tasks,":[85],"collects":[86],"rollouts,":[88],"derives":[90],"trajectory-level":[91,140],"supervision":[92,179],"model":[95],"training.":[96],"As":[97],"new":[98],"threats":[99],"validation":[101],"failures":[102],"appear,":[103],"the":[104,136,170],"pipeline":[105],"can":[106,189],"be":[107],"repeated,":[108],"yielding":[109],"an":[110],"adaptive":[111,206],"loop":[113],"rather":[114],"than":[115],"static,":[117],"benchmark-driven":[118],"process.":[120],"instantiate":[122],"by":[124],"multiple":[126],"backbones,":[128],"including":[129],"Qwen3-Guard":[130],"Llama-Guard":[132],"variants,":[133],"evaluate":[135],"resulting":[137],"guards":[138],"on":[139],"agent-safety":[141],"benchmarks.":[142],"consistently":[144],"improves":[145,155],"detection":[147,156],"across":[148],"On":[151],"AgentHazard,":[152],"it":[153],"substantially":[154],"accuracy":[157,163],"over":[158],"off-the-shelf":[159],"models,":[161],"increasing":[164],"38.79%":[166],"82.38%":[168],"under":[169],"averaged":[171],"guard-model":[172],"setting.":[173],"These":[174],"results":[175],"show":[176],"grounded":[180],"in":[181],"discovery":[184],"improve":[190],"monitoring":[192],"beyond":[193],"fixed":[194],"taxonomies":[195],"synthetic":[197],"prompt-level":[198],"data.":[199],"offers":[201],"scalable":[203],"path":[204],"toward":[205],"defenses":[207],"facing":[211],"evolving":[212],"real-world":[213],"risks.":[214]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}
