{"id":"https://openalex.org/W4388329043","doi":"https://doi.org/10.48550/arxiv.2311.01011","title":"Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game","display_name":"Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game","publication_year":2023,"publication_date":"2023-11-02","ids":{"openalex":"https://openalex.org/W4388329043","doi":"https://doi.org/10.48550/arxiv.2311.01011"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2311.01011","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.01011","pdf_url":"https://arxiv.org/pdf/2311.01011","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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/2311.01011","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5085124552","display_name":"Sam Toyer","orcid":"https://orcid.org/0000-0002-6665-6593","observed_orcids":["https://orcid.org/0000-0002-6665-6593"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Toyer, Sam","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058976886","display_name":"Olivia Watkins","orcid":"https://orcid.org/0000-0003-3054-0385","observed_orcids":["https://orcid.org/0000-0003-3054-0385"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Watkins, Olivia","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036734649","display_name":"Ethan Mendes","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mendes, Ethan Adrian","raw_affiliation_strings":["Georgia Tech","UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Tech","institution_ids":["https://openalex.org/I130701444"]},{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080873677","display_name":"Justin Svegliato","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Svegliato, Justin","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011630384","display_name":"Luke Bailey","orcid":"https://orcid.org/0009-0000-7257-9507","observed_orcids":["https://orcid.org/0009-0000-7257-9507","https://orcid.org/0000-0003-2213-7430"]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bailey, Luke","raw_affiliation_strings":["Harvard University","UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University","institution_ids":["https://openalex.org/I136199984"]},{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Wang, Tiffany","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wang, Tiffany","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066782748","display_name":"Isaac Ong","orcid":"https://orcid.org/0009-0006-3551-7394","observed_orcids":["https://orcid.org/0009-0006-3551-7394"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ong, Isaac","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064150422","display_name":"Karim Elmaaroufi","orcid":"https://orcid.org/0009-0002-7421-8443","observed_orcids":["https://orcid.org/0009-0002-7421-8443"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Elmaaroufi, Karim","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049349154","display_name":"Pieter Abbeel","orcid":"https://orcid.org/0009-0007-0628-9932","observed_orcids":["https://orcid.org/0009-0007-0628-9932"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abbeel, Pieter","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029105520","display_name":"Trevor Darrell","orcid":"https://orcid.org/0000-0001-5453-8533","observed_orcids":["https://orcid.org/0000-0001-5453-8533"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Darrell, Trevor","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039096905","display_name":"Alan Ritter","orcid":"https://orcid.org/0009-0001-4602-138X","observed_orcids":["https://orcid.org/0009-0001-4602-138X"]},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ritter, Alan","raw_affiliation_strings":["Georgia Tech"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Tech","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007305440","display_name":"Stuart Russell","orcid":"https://orcid.org/0000-0001-5252-4306","observed_orcids":["https://orcid.org/0000-0001-5252-4306"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Russell, Stuart","raw_affiliation_strings":["UC Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley","institution_ids":["https://openalex.org/I95457486"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":106,"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":1.0,"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":1.0,"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/T10028","display_name":"Topic Modeling","score":0.0,"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/T11424","display_name":"Security and Verification in Computing","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/prompt-injection","display_name":"prompt injection","score":0.9909999966621399},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8930000066757202},{"id":"https://openalex.org/keywords/adversarial-examples","display_name":"adversarial examples","score":0.9390000104904175},{"id":"https://openalex.org/keywords/interpretability","display_name":"interpretability","score":0.5910000205039978},{"id":"https://openalex.org/keywords/online-games","display_name":"online games","score":0.5759999752044678},{"id":"https://openalex.org/keywords/llm-security","display_name":"LLM security","score":0.6010000109672546},{"id":"https://openalex.org/keywords/attack-dataset","display_name":"attack dataset","score":0.46000000834465027}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.7941982746124268},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.6909013390541077},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6573191285133362},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.6420297026634216},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.6096648573875427},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.5504041910171509},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.5315859913825989},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.40076953172683716},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3827097415924072},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.37182334065437317},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34155550599098206},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.09819114208221436},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.07619211077690125},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2311.01011","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.01011","pdf_url":"https://arxiv.org/pdf/2311.01011","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2311.01011","doi":"https://doi.org/10.48550/arxiv.2311.01011","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2311.01011","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2311.01011","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.01011","pdf_url":"https://arxiv.org/pdf/2311.01011","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G2775513529","display_name":"Secure, Real-Time Decisions on Live Data","funder_award_id":"1730628","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320332169","display_name":"Directorate for Computer and Information Science and Engineering","ror":"https://ror.org/025kzpk63"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4388329043.pdf","grobid_xml":"https://content.openalex.org/works/W4388329043.grobid-xml"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W2970290563","https://openalex.org/W4309395891","https://openalex.org/W4311991106","https://openalex.org/W4327810158","https://openalex.org/W4377121468","https://openalex.org/W4380353722","https://openalex.org/W4384648205","https://openalex.org/W4388886073","https://openalex.org/W4391724817","https://openalex.org/W4405181600","https://openalex.org/W7133204634"],"related_works":["https://openalex.org/W7147211113","https://openalex.org/W7127370637","https://openalex.org/W4389157312","https://openalex.org/W4417308938","https://openalex.org/W7166981937","https://openalex.org/W4417261870","https://openalex.org/W7131834642","https://openalex.org/W7223042973","https://openalex.org/W4411337807","https://openalex.org/W4402217583"],"abstract_inverted_index":{"While":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"are":[5,138],"increasingly":[6],"being":[7],"used":[8],"in":[9,86,144],"real-world":[10],"applications,":[11,163],"they":[12,166],"remain":[13],"vulnerable":[14,139],"to":[15,109,115,124,140,160,174],"prompt":[16,44,52,119,126,129],"injection":[17,45],"attacks:":[18],"malicious":[19],"third":[20],"party":[21],"prompts":[22],"that":[23,135,152],"subvert":[24],"the":[25,28,66,74,100,107,141,145,157,175],"intent":[26],"of":[27,41,58,68,77,92,102,118,172],"system":[29],"designer.":[30],"To":[31,65],"help":[32],"researchers":[33],"study":[34],"this":[35,71],"problem,":[36],"we":[37,122,150],"present":[38],"a":[39,90,111,168],"dataset":[40,76,88,108,158],"over":[42],"126,000":[43],"attacks":[46,85],"and":[47,96,128,181],"46,000":[48],"prompt-based":[49],"\"defenses\"":[50],"against":[51],"injection,":[53,120],"all":[54,179],"created":[55],"by":[56],"players":[57],"an":[59],"online":[60],"game":[61],"called":[62],"Tensor":[63,146],"Trust.":[64],"best":[67],"our":[69,87],"knowledge,":[70],"is":[72],"currently":[73],"largest":[75],"human-generated":[78],"adversarial":[79],"examples":[80],"for":[81,113],"instruction-following":[82],"LLMs.":[83,103],"The":[84],"have":[89,167],"lot":[91],"easily":[93],"interpretable":[94],"stucture,":[95],"shed":[97],"light":[98],"on":[99],"weaknesses":[101],"We":[104,177],"also":[105],"use":[106],"create":[110],"benchmark":[112,132],"resistance":[114],"two":[116],"types":[117],"which":[121],"refer":[123],"as":[125],"extraction":[127],"hijacking.":[130],"Our":[131],"results":[133],"show":[134,151],"many":[136],"models":[137],"attack":[142,154],"strategies":[143,155],"Trust":[147],"dataset.":[148],"Furthermore,":[149],"some":[153],"from":[156],"generalize":[159],"deployed":[161],"LLM-based":[162],"even":[164],"though":[165],"very":[169],"different":[170],"set":[171],"constraints":[173],"game.":[176],"release":[178],"data":[180],"source":[182],"code":[183],"at":[184],"https://tensortrust.ai/paper":[185]},"counts_by_year":[{"year":2026,"cited_by_count":45},{"year":2025,"cited_by_count":28},{"year":2024,"cited_by_count":31},{"year":2023,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2023-11-04T00:00:00"}