{"id":"https://openalex.org/W4412888061","doi":"https://doi.org/10.18653/v1/2025.findings-acl.784","title":"Enhancing LLM Agent Safety via Causal Influence Prompting","display_name":"Enhancing LLM Agent Safety via Causal Influence Prompting","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888061","doi":"https://doi.org/10.18653/v1/2025.findings-acl.784"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.784","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.784","pdf_url":"https://aclanthology.org/2025.findings-acl.784.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 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.784.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Dongyoon Hahm","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"Dongyoon Hahm","raw_affiliation_strings":["KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I4210099236","https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Woogyeol Jin","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"Woogyeol Jin","raw_affiliation_strings":["KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I4210099236","https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"June Suk Choi","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"June Suk Choi","raw_affiliation_strings":["KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I4210099236","https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005535132","display_name":"Sungsoo Ahn","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"Sungsoo Ahn","raw_affiliation_strings":["KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I4210099236","https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033571780","display_name":"Kimin Lee","orcid":"https://orcid.org/0000-0001-9017-3084"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"Kimin Lee","raw_affiliation_strings":["KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I4210099236","https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15143","last_page":"15168"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10260","display_name":"Software Engineering Research","score":0.9514999985694885,"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"}},"topics":[{"id":"https://openalex.org/T10260","display_name":"Software Engineering Research","score":0.9514999985694885,"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"}},{"id":"https://openalex.org/T10743","display_name":"Software Testing and Debugging Techniques","score":0.9488999843597412,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/T13295","display_name":"Safety Systems Engineering in Autonomy","score":0.946399986743927,"subfield":{"id":"https://openalex.org/subfields/2213","display_name":"Safety, Risk, Reliability and Quality"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5670208930969238},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.3517891466617584},{"id":"https://openalex.org/keywords/risk-analysis","display_name":"Risk analysis (engineering)","score":0.3259803056716919},{"id":"https://openalex.org/keywords/business","display_name":"Business","score":0.23800116777420044}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5670208930969238},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3517891466617584},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.3259803056716919},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.23800116777420044}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.784","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.784","pdf_url":"https://aclanthology.org/2025.findings-acl.784.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 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.784","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.784","pdf_url":"https://aclanthology.org/2025.findings-acl.784.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 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3736587842","display_name":null,"funder_award_id":"RS-2019-II190075","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G4468082379","display_name":null,"funder_award_id":"RS-2019-II190075","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"}],"funders":[{"id":"https://openalex.org/F4320324161","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"},{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888061.pdf","grobid_xml":"https://content.openalex.org/works/W4412888061.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"As":[0],"autonomous":[1],"agents":[2,60],"powered":[3],"by":[4],"large":[5],"language":[6],"models":[7],"(LLMs)":[8],"continue":[9],"to":[10,43,61,83,140],"demonstrate":[11,111],"potential":[12],"across":[13],"various":[14,235],"assistive":[15],"tasks,":[16],"ensuring":[17],"their":[18],"safe":[19],"and":[20,45,65,98,108,122,224],"reliable":[21],"behavior":[22],"is":[23],"crucial":[24],"for":[25],"preventing":[26],"unintended":[27],"consequences.In":[28],"this":[29],"work,":[30],"we":[31],"introduce":[32],"CIP,":[33],"a":[34,53,77,147],"novel":[35],"technique":[36],"that":[37,112,145],"leverages":[38],"causal":[39],"influence":[40],"diagrams":[41],"(CIDs)":[42],"identify":[44],"mitigate":[46],"risks":[47],"arising":[48],"from":[49],"agent":[50,90],"decision-making.CIDs":[51],"provide":[52],"structured":[54],"representation":[55],"of":[56,71,156],"cause-and-effect":[57],"relationships,":[58],"enabling":[59],"anticipate":[62],"harmful":[63],"outcomes":[64],"make":[66],"safer":[67],"decisions.Our":[68],"approach":[69],"consists":[70],"three":[72],"key":[73],"steps:":[74],"(1)":[75],"initializing":[76],"CID":[78,103,127,163,217,219,225,253],"based":[79,104],"on":[80,105],"task":[81],"specifications":[82],"outline":[84],"the":[85,93,96,102,136,153,162,165,231],"decision-making":[86,154],"process,":[87],"(2)":[88],"guiding":[89],"interactions":[91],"with":[92,234],"environment":[94],"using":[95,164],"CID,":[97],"(3)":[99],"iteratively":[100],"refining":[101],"observed":[106],"behaviors":[107],"outcomes.Experimental":[109],"results":[110],"our":[113],"method":[114],"effectively":[115],"enhances":[116],"safety":[117],"in":[118],"both":[119],"code":[120],"execution":[121,223],"mobile":[123],"device":[124],"control":[125],"tasks.":[126],"Generation":[128],"PromptAgent":[129],"action":[130],"space:":[131],"swipe":[132],"Task":[133,222],"instruction:":[134],"Forward":[135],"most":[137],"recent":[138],"message":[139],"John.You":[141],"are":[142],"an":[143,157],"AI":[144],"generates":[146],"Causal":[148],"Influence":[149],"Diagram":[150],"(CID)":[151],"representing":[152],"process":[155],"LLM":[158],"agent.(...)":[159],"Finally,":[160],"construct":[161],"provided":[166],"functions.":[167],"Large":[168],"Language":[169],"Model$":[170],"add_node(C1,":[171],"\"User":[172],"Instruction\")":[173],"#":[174,183,190,197,204],"Chance":[175,184],"node":[176,185,192,199,206],"$":[177,186,193,200,207,243],"add_node(C2,":[178],"\"Identify":[179],"Most":[180],"Recent":[181],"Message\")":[182,189],"add_node(D1,":[187],"\"Forward":[188],"Decision":[191],"add_node(U1,":[194],"\"Successful":[195],"Forwarding\")":[196],"Utility":[198,205],"add_node(U2,":[201],"\"Privacy":[202],"Violation\")":[203],"add_edge(C1,":[208],"C2);":[209],"add_edge(C2,":[210,244],"D1);":[211],"add_edge(D1,":[212,214],"U1);":[213],"U2)":[215],"Generated":[216],"(a)":[218],"generation":[220],"(b)":[221],"refinement":[226,254],"Agent":[227],"The":[228],"screen":[229,233],"shows":[230],"home":[232],"app":[236],"icons":[237],"including":[238],"Messages.$":[239],"add_node(C3,":[240],"\"Consent":[241],"Requirement\")":[242],"C3);":[245],"add_edge(C3,":[246],"D1)":[247],"tap":[248],"input":[249],"text":[250],"refuse":[251],"...":[252]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}