{"id":"https://openalex.org/W7128714846","doi":"https://doi.org/10.48448/ytbt-6r06","title":"460 - Towards Effective Offensive Security LLM Agents: Hyperparameter Tuning, LLM as a Judge, and a Lightweight CTF Benchmark","display_name":"460 - Towards Effective Offensive Security LLM Agents: Hyperparameter Tuning, LLM as a Judge, and a Lightweight CTF Benchmark","publication_year":2026,"publication_date":"2026-01-07","ids":{"openalex":"https://openalex.org/W7128714846","doi":"https://doi.org/10.48448/ytbt-6r06"},"language":"en","primary_location":{"id":"doi:10.48448/ytbt-6r06","doi":"https://doi.org/10.48448/ytbt-6r06","is_oa":true,"landing_page_url":"https://doi.org/10.48448/ytbt-6r06","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"volume":null,"type":"other","is_traditional":false,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48448/ytbt-6r06","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125741655","display_name":"Association for Artificial Intelligence 2026","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Association for Artificial Intelligence 2026","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063578033","display_name":"Saksham Aggarwal","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aggarwal, Saksham","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125759002","display_name":"Ramesh Karri","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Karri, Ramesh","raw_affiliation_strings":["NYU"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NYU","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082413942","display_name":"Farshad Khorrami","orcid":"https://orcid.org/0000-0002-8418-004X","observed_orcids":["https://orcid.org/0000-0002-8418-004X"]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Khorrami, Farshad","raw_affiliation_strings":["New York University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054769060","display_name":"P. Krishnamurthy","orcid":"https://orcid.org/0000-0001-8264-7972","observed_orcids":["https://orcid.org/0000-0001-8264-7972"]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Krishnamurthy, Prashanth","raw_affiliation_strings":["New York University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099097255","display_name":"Kimberly Milner","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Milner, Kimberly","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068986465","display_name":"Venkata Sai Charan Putrevu","orcid":"https://orcid.org/0000-0003-4535-8690","observed_orcids":["https://orcid.org/0000-0003-4535-8690"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Putrevu, Venkata","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001530440","display_name":"Nanda Rani","orcid":"https://orcid.org/0000-0003-1255-5284","observed_orcids":["https://orcid.org/0000-0003-1255-5284"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rani, Nanda","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125719481","display_name":"Muhammad Shafique","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shafique, Muhammad","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113123602","display_name":"Minghao Shao","orcid":"https://orcid.org/0009-0002-4467-6224","observed_orcids":["https://orcid.org/0009-0002-4467-6224","https://orcid.org/0009-0003-6060-5693"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shao, Minghao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125509541","display_name":"Sandeep K. Shukla","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shukla, Sandeep","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007457263","display_name":"Meet Udeshi","orcid":"https://orcid.org/0000-0001-7297-0880","observed_orcids":["https://orcid.org/0000-0001-7297-0880"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Udeshi, Meet","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5101377025","display_name":"Haoran Xi","orcid":"https://orcid.org/0009-0005-3268-7366","observed_orcids":["https://orcid.org/0009-0005-3268-7366","https://orcid.org/0009-0009-3051-1630"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xi, Haoran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":0.4918999969959259,"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":0.4918999969959259,"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/T10734","display_name":"Information and Cyber Security","score":0.21549999713897705,"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/T11424","display_name":"Security and Verification in Computing","score":0.11649999767541885,"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/llm-based-agents","display_name":"LLM agents","score":0.9089999794960022},{"id":"https://openalex.org/keywords/offensive-security","display_name":"offensive security","score":0.984000027179718},{"id":"https://openalex.org/keywords/capture-the-flag","display_name":"Capture the Flag","score":0.7390000224113464},{"id":"https://openalex.org/keywords/llm-as-a-judge","display_name":"LLM-as-a-judge","score":0.7269999980926514},{"id":"https://openalex.org/keywords/hyperparameter-optimization","display_name":"hyperparameter optimization","score":0.8560000061988831},{"id":"https://openalex.org/keywords/temperature","display_name":"temperature","score":0.7369999885559082},{"id":"https://openalex.org/keywords/top-p-sampling","display_name":"top-p sampling","score":0.8870000243186951},{"id":"https://openalex.org/keywords/multiagent-coordination","display_name":"multi-agent coordination","score":0.6940000057220459},{"id":"https://openalex.org/keywords/cryptography","display_name":"cryptography","score":0.367000013589859},{"id":"https://openalex.org/keywords/reverse-engineering","display_name":"reverse engineering","score":0.6880000233650208},{"id":"https://openalex.org/keywords/forensic-science","display_name":"forensic science","score":0.7229999899864197},{"id":"https://openalex.org/keywords/web-security","display_name":"web security","score":0.6110000014305115}],"concepts":[{"id":"https://openalex.org/C176856949","wikidata":"https://www.wikidata.org/wiki/Q2001676","display_name":"Offensive","level":2,"score":0.7767000198364258},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.7175999879837036},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6600000262260437},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.46239998936653137},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.44020000100135803},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.43790000677108765},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.40149998664855957},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.3977999985218048},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.3903000056743622},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37229999899864197},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.36629998683929443},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3569999933242798},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3239000141620636},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.2791000008583069},{"id":"https://openalex.org/C2777548347","wikidata":"https://www.wikidata.org/wiki/Q5456937","display_name":"Flagging","level":2,"score":0.2757999897003174},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.2639999985694885},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.26269999146461487},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.25699999928474426}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48448/ytbt-6r06","doi":"https://doi.org/10.48448/ytbt-6r06","is_oa":true,"landing_page_url":"https://doi.org/10.48448/ytbt-6r06","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"}],"best_oa_location":{"id":"doi:10.48448/ytbt-6r06","doi":"https://doi.org/10.48448/ytbt-6r06","is_oa":true,"landing_page_url":"https://doi.org/10.48448/ytbt-6r06","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7128719387","https://openalex.org/W7162107299","https://openalex.org/W7161035179","https://openalex.org/W4407687097","https://openalex.org/W7128506643","https://openalex.org/W4391987669","https://openalex.org/W4415883730","https://openalex.org/W7155451781","https://openalex.org/W7155246080","https://openalex.org/W7130581202"],"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":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-02-13T00:00:00"}