{"id":"https://openalex.org/W7214760947","doi":"https://doi.org/10.48550/arxiv.2609.34415","title":"PROACT-Agent: Progressive Runtime Oversight and Active Circuit-breaking for Real-Time Safety","display_name":"PROACT-Agent: Progressive Runtime Oversight and Active Circuit-breaking for Real-Time Safety","publication_year":2026,"publication_date":"2026-09-28","ids":{"openalex":"https://openalex.org/W7214760947","doi":"https://doi.org/10.48550/arxiv.2609.34415"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.34415","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.34415","pdf_url":"https://arxiv.org/pdf/2609.34415","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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.34415","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101644889","display_name":"Ding Jia","orcid":"https://orcid.org/0000-0002-1824-0710","observed_orcids":["https://orcid.org/0000-0002-1824-0710"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia, Ding","raw_affiliation_strings":["State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153434848","display_name":"Wei Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Wei","raw_affiliation_strings":["State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153538846","display_name":"Xianglong Du","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Du, Xianglong","raw_affiliation_strings":["State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153425612","display_name":"Yingjie Li","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yingjie","raw_affiliation_strings":["State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107299395","display_name":"Yi Yang","orcid":"https://orcid.org/0000-0002-9024-4150","observed_orcids":["https://orcid.org/0000-0002-9024-4150"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Yingqing","raw_affiliation_strings":["State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153457712","display_name":"Huili Yu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Huili","raw_affiliation_strings":["State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075431520","display_name":"\u8a79\u957f\u677e","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhan, Zhangsong","raw_affiliation_strings":["State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Intelligent Vehicle Safety Technology , Changan Automobile","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5112610083","display_name":"Chu Zhou","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Chu","raw_affiliation_strings":["Independent Researcher"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Independent Researcher","institution_ids":[]}]}],"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":true,"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":0.9955999851226807,"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":0.9955999851226807,"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":0.0012000000569969416,"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":0.0005000000237487257,"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/llm-based-agents","display_name":"LLM agents","score":0.7829999923706055},{"id":"https://openalex.org/keywords/circuit-breaking","display_name":"circuit breaking","score":0.640999972820282},{"id":"https://openalex.org/keywords/causal-consistency","display_name":"causal consistency","score":0.9210000038146973},{"id":"https://openalex.org/keywords/safety-benchmarking","display_name":"safety benchmarking","score":0.5429999828338623},{"id":"https://openalex.org/keywords/agentdojo","display_name":"AgentDojo","score":0.5329999923706055},{"id":"https://openalex.org/keywords/denial-of-service-attacks","display_name":"denial-of-service attacks","score":0.5559999942779541}],"concepts":[{"id":"https://openalex.org/C141141315","wikidata":"https://www.wikidata.org/wiki/Q2379942","display_name":"Guard (computer science)","level":2,"score":0.715499997138977},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6858000159263611},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6223999857902527},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5234000086784363},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.45989999175071716},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.45910000801086426},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.40380001068115234},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.384799987077713},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.3781000077724457},{"id":"https://openalex.org/C50942859","wikidata":"https://www.wikidata.org/wiki/Q4967193","display_name":"Rectification","level":3,"score":0.3603000044822693},{"id":"https://openalex.org/C2780665704","wikidata":"https://www.wikidata.org/wiki/Q959298","display_name":"Intervention (counseling)","level":2,"score":0.34950000047683716},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33980000019073486},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2741999924182892},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.2590999901294708}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.34415","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.34415","pdf_url":"https://arxiv.org/pdf/2609.34415","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.2609.34415","doi":"https://doi.org/10.48550/arxiv.2609.34415","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.34415","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.34415","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.34415","pdf_url":"https://arxiv.org/pdf/2609.34415","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"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214760947.pdf","grobid_xml":"https://content.openalex.org/works/W7214760947.grobid-xml"},"referenced_works_count":25,"referenced_works":["https://openalex.org/W2518108298","https://openalex.org/W2908510526","https://openalex.org/W4389713766","https://openalex.org/W4401306838","https://openalex.org/W4405627542","https://openalex.org/W4405655184","https://openalex.org/W4414281281","https://openalex.org/W4414939355","https://openalex.org/W4415333036","https://openalex.org/W4416034196","https://openalex.org/W4416071203","https://openalex.org/W4416116727","https://openalex.org/W4416608657","https://openalex.org/W4416942889","https://openalex.org/W4417006854","https://openalex.org/W4417068164","https://openalex.org/W4417087067","https://openalex.org/W7090866534","https://openalex.org/W7092296780","https://openalex.org/W7115592956","https://openalex.org/W7125768537","https://openalex.org/W7127380112","https://openalex.org/W7128378279","https://openalex.org/W7130402983","https://openalex.org/W7166886473"],"related_works":["https://openalex.org/W4416551460","https://openalex.org/W7201884335","https://openalex.org/W7214238282","https://openalex.org/W7155077488","https://openalex.org/W7118014120","https://openalex.org/W7166886473","https://openalex.org/W4399695705","https://openalex.org/W7223462612","https://openalex.org/W7212001064","https://openalex.org/W7090866534"],"abstract_inverted_index":{"The":[0],"transition":[1],"from":[2,12,145],"Large":[3],"Language":[4],"Models":[5],"(LLMs)":[6],"to":[7,15,46,64,76,87,147],"agents":[8],"shifts":[9],"safety":[10,105],"stakes":[11],"toxic":[13],"text":[14],"irreversible":[16],"environmental":[17],"harm.":[18],"While":[19],"current":[20],"defenses":[21],"remain":[22],"largely":[23],"retrospective,":[24],"proactive":[25],"runtime":[26],"intervention":[27],"is":[28],"bottlenecked":[29],"by":[30],"the":[31,118,122],"lack":[32],"of":[33],"large-scale,":[34],"causally-consistent":[35],"data.":[36],"We":[37,50,100],"propose":[38],"PROACT-Agent,":[39],"a":[40,52,103],"framework":[41],"for":[42,97],"synthesizing":[43],"high-fidelity":[44],"trajectories":[45],"enable":[47],"real-time":[48],"guardrails.":[49],"identify":[51],"critical":[53],"\"safety":[54],"drift\"":[55],"in":[56,80],"prior":[57],"benchmarks,":[58],"where":[59],"lenient":[60],"annotation":[61],"paradigms":[62],"fail":[63],"enforce":[65,88],"temporal":[66],"consistency.":[67],"PROACT-Agent":[68],"addresses":[69],"this":[70],"through:":[71],"(1)":[72],"Progressive":[73],"Trajectory":[74],"Unrolling":[75],"reveal":[77],"risks":[78],"hidden":[79],"long-context":[81],"interactions;":[82],"(2)":[83],"Reasoning-Augmented":[84],"Causal":[85],"Rectification":[86],"monotonic":[89],"causal":[90],"consistency;":[91],"and":[92,129],"(3)":[93],"Culturally-Aware":[94],"Data":[95],"Localization":[96],"cross-border":[98],"robustness.":[99],"introduce":[101],"PROACT-Bench,":[102],"bilingual":[104],"benchmark":[106],"with":[107],"155,780":[108],"states":[109],"labeled":[110],"through":[111],"multi-model":[112],"adjudication.":[113],"Evaluating":[114],"updated":[115],"context":[116],"before":[117],"next":[119],"LLM":[120],"inference,":[121],"trained":[123],"guard":[124],"achieves":[125],"91.46%":[126],"unsafe-class":[127],"F1":[128],"90.63%":[130],"exact-boundary":[131],"detection":[132],"under":[133],"complete":[134],"source":[135],"holdout.":[136],"In":[137],"AgentDojo,":[138],"it":[139],"reduces":[140],"non-DoS":[141],"targeted":[142],"attack":[143],"success":[144],"20.82%":[146],"0.40%.":[148]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-30T00:00:00"}