{"id":"https://openalex.org/W7138370235","doi":"https://doi.org/10.1609/aaai.v40i35.40210","title":"Towards Effective Offensive Security LLM Agents: Hyperparameter Tuning, LLM as a Judge, and a Lightweight CTF Benchmark","display_name":"Towards Effective Offensive Security LLM Agents: Hyperparameter Tuning, LLM as a Judge, and a Lightweight CTF Benchmark","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138370235","doi":"https://doi.org/10.1609/aaai.v40i35.40210"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i35.40210","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i35.40210","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i35.40210","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129668368","display_name":"Minghao Shao","orcid":null},"institutions":[{"id":"https://openalex.org/I120250893","display_name":"New York University Abu Dhabi","ror":"https://ror.org/00e5k0821","country_code":"AE","type":"education","lineage":["https://openalex.org/I120250893","https://openalex.org/I57206974"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Minghao Shao","raw_affiliation_strings":["NYU Tandon School of Engineering\nNYU Abu Dhabi"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering\nNYU Abu Dhabi","institution_ids":["https://openalex.org/I120250893"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001530440","display_name":"Nanda Rani","orcid":null},"institutions":[{"id":"https://openalex.org/I94234084","display_name":"Indian Institute of Technology Kanpur","ror":"https://ror.org/05pjsgx75","country_code":"IN","type":"education","lineage":["https://openalex.org/I94234084"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Nanda Rani","raw_affiliation_strings":["Indian Institute of Technology Kanpur"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Indian Institute of Technology Kanpur","institution_ids":["https://openalex.org/I94234084"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099097255","display_name":"Kimberly Milner","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kimberly Milner","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101377025","display_name":"Haoran Xi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haoran Xi","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007457263","display_name":"Meet Udeshi","orcid":"https://orcid.org/0000-0001-7297-0880"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meet Udeshi","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063578033","display_name":"Saksham Aggarwal","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Saksham Aggarwal","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068986465","display_name":"Venkata Sai Charan Putrevu","orcid":"https://orcid.org/0000-0003-3495-8332"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Venkata Sai Charan Putrevu","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129674390","display_name":"Sandeep K. Shukla","orcid":null},"institutions":[{"id":"https://openalex.org/I64189192","display_name":"International Institute of Information Technology, Hyderabad","ror":"https://ror.org/05f11g639","country_code":"IN","type":"education","lineage":["https://openalex.org/I64189192"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sandeep K. Shukla","raw_affiliation_strings":["International Institute of Information Technology Hyderabad"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"International Institute of Information Technology Hyderabad","institution_ids":["https://openalex.org/I64189192"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129668776","display_name":"Prashanth Krishnamurthy","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Prashanth Krishnamurthy","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082413942","display_name":"Farshad Khorrami","orcid":"https://orcid.org/0000-0002-8418-004X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Farshad Khorrami","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129731607","display_name":"Ramesh Karri","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ramesh Karri","raw_affiliation_strings":["NYU Tandon School of Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Tandon School of Engineering","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129684384","display_name":"Muhammad Shafique","orcid":null},"institutions":[{"id":"https://openalex.org/I120250893","display_name":"New York University Abu Dhabi","ror":"https://ror.org/00e5k0821","country_code":"AE","type":"education","lineage":["https://openalex.org/I120250893","https://openalex.org/I57206974"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Muhammad Shafique","raw_affiliation_strings":["NYU Abu Dhabi"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU Abu Dhabi","institution_ids":["https://openalex.org/I120250893"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"35","first_page":"29660","last_page":"29668"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11241","display_name":"Advanced Malware Detection Techniques","score":0.3871999979019165,"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"}},"topics":[{"id":"https://openalex.org/T11241","display_name":"Advanced Malware Detection Techniques","score":0.3871999979019165,"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/T10400","display_name":"Network Security and Intrusion Detection","score":0.1266999989748001,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.06599999964237213,"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/offensive","display_name":"Offensive","score":0.7631999850273132},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.6793000102043152},{"id":"https://openalex.org/keywords/automation","display_name":"Automation","score":0.48829999566078186},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.46129998564720154},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4472000002861023},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.39660000801086426},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.37709999084472656},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.37450000643730164}],"concepts":[{"id":"https://openalex.org/C176856949","wikidata":"https://www.wikidata.org/wiki/Q2001676","display_name":"Offensive","level":2,"score":0.7631999850273132},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6793000102043152},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6646000146865845},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.48829999566078186},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.46129998564720154},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4472000002861023},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.413100004196167},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.39660000801086426},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.37709999084472656},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.37450000643730164},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35179999470710754},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.3463999927043915},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32089999318122864},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3206999897956848},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.2930999994277954},{"id":"https://openalex.org/C2777548347","wikidata":"https://www.wikidata.org/wiki/Q5456937","display_name":"Flagging","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.2728999853134155},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.2702000141143799},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C121822524","wikidata":"https://www.wikidata.org/wiki/Q5157582","display_name":"Computer security model","level":2,"score":0.2540000081062317},{"id":"https://openalex.org/C2780264999","wikidata":"https://www.wikidata.org/wiki/Q7445032","display_name":"Security domain","level":2,"score":0.2524999976158142},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.2502000033855438}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i35.40210","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i35.40210","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i35.40210","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i35.40210","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.5207609534263611,"id":"https://metadata.un.org/sdg/16","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":{"Recent":[0],"advances":[1],"in":[2,149],"LLM":[3,50,93,146],"agentic":[4],"systems":[5],"have":[6],"improved":[7],"the":[8,17,24,142],"automation":[9],"of":[10,119],"offensive":[11,40],"security":[12,41],"tasks,":[13],"particularly":[14],"for":[15,36,76,144],"Capture":[16],"Flag":[18],"(CTF)":[19],"challenges.":[20],"We":[21],"systematically":[22],"investigate":[23],"key":[25],"factors":[26],"that":[27],"drive":[28],"agent":[29,56,82,103,147],"success":[30],"and":[31,58,98,105,131,140],"provide":[32,59],"a":[33,47,52,69,116],"detailed":[34],"recipe":[35],"building":[37],"effective":[38],"LLM-based":[39],"agents.":[42],"First,":[43],"we":[44,67,90,113],"present":[45,114],"CTFJudge,":[46],"framework":[48],"leveraging":[49],"as":[51],"judge":[53],"to":[54],"analyze":[55],"trajectories":[57],"granular":[60],"evaluation":[61],"across":[62,124],"CTF":[63,72,122],"solving":[64],"steps.":[65],"Second,":[66],"propose":[68],"novel":[70],"metric,":[71],"Competency":[73],"Index":[74],"(CCI)":[75],"partial":[77],"correctness,":[78],"revealing":[79],"how":[80,92],"closely":[81],"solutions":[83],"align":[84],"with":[85],"human-crafted":[86],"gold":[87],"standards.":[88],"Third,":[89],"examine":[91],"hyperparameters,":[94],"namely":[95],"temperature,":[96],"top-p,":[97],"maximum":[99],"token":[100],"length,":[101],"influence":[102],"performance":[104],"automated":[106],"cybersecurity":[107],"task":[108],"planning.":[109],"For":[110],"rapid":[111],"evaluation,":[112],"CTFTiny,":[115],"curated":[117],"benchmark":[118],"50":[120],"representative":[121],"challenges":[123],"binary":[125],"exploitation,":[126],"web,":[127],"reverse":[128],"engineering,":[129],"forensics,":[130],"cryptography.":[132],"Our":[133],"findings":[134],"identify":[135],"optimal":[136],"multi-agent":[137],"coordination":[138],"settings":[139],"lay":[141],"groundwork":[143],"future":[145],"research":[148],"cybersecurity.":[150]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-03-18T00:00:00"}
