{"id":"https://openalex.org/W4416251571","doi":"https://doi.org/10.1109/ijcnn64981.2025.11227336","title":"Adversarial Attack on Large Language Models using Exponentiated Gradient Descent","display_name":"Adversarial Attack on Large Language Models using Exponentiated Gradient Descent","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4416251571","doi":"https://doi.org/10.1109/ijcnn64981.2025.11227336"},"language":null,"primary_location":{"id":"doi:10.1109/ijcnn64981.2025.11227336","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11227336","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114244909","display_name":"Sajib Biswas","orcid":null},"institutions":[{"id":"https://openalex.org/I103163165","display_name":"Florida State University","ror":"https://ror.org/05g3dte14","country_code":"US","type":"education","lineage":["https://openalex.org/I103163165"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Sajib Biswas","raw_affiliation_strings":["Florida State University,Department of Computer Science,Tallahassee,Florida"],"affiliations":[{"raw_affiliation_string":"Florida State University,Department of Computer Science,Tallahassee,Florida","institution_ids":["https://openalex.org/I103163165"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110145764","display_name":"Mao Nishino","orcid":null},"institutions":[{"id":"https://openalex.org/I103163165","display_name":"Florida State University","ror":"https://ror.org/05g3dte14","country_code":"US","type":"education","lineage":["https://openalex.org/I103163165"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mao Nishino","raw_affiliation_strings":["Florida State University,Department of Mathematics,Tallahassee,Florida"],"affiliations":[{"raw_affiliation_string":"Florida State University,Department of Mathematics,Tallahassee,Florida","institution_ids":["https://openalex.org/I103163165"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033842275","display_name":"Samuel Jacob Chacko","orcid":null},"institutions":[{"id":"https://openalex.org/I103163165","display_name":"Florida State University","ror":"https://ror.org/05g3dte14","country_code":"US","type":"education","lineage":["https://openalex.org/I103163165"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Samuel Jacob Chacko","raw_affiliation_strings":["Florida State University,Department of Computer Science,Tallahassee,Florida"],"affiliations":[{"raw_affiliation_string":"Florida State University,Department of Computer Science,Tallahassee,Florida","institution_ids":["https://openalex.org/I103163165"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5025749256","display_name":"Xiuwen Liu","orcid":"https://orcid.org/0000-0002-6202-1937"},"institutions":[{"id":"https://openalex.org/I103163165","display_name":"Florida State University","ror":"https://ror.org/05g3dte14","country_code":"US","type":"education","lineage":["https://openalex.org/I103163165"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiuwen Liu","raw_affiliation_strings":["Florida State University,Department of Computer Science,Tallahassee,Florida"],"affiliations":[{"raw_affiliation_string":"Florida State University,Department of Computer Science,Tallahassee,Florida","institution_ids":["https://openalex.org/I103163165"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5114244909"],"corresponding_institution_ids":["https://openalex.org/I103163165"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.19480257,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.6152999997138977,"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":0.6152999997138977,"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":0.06369999796152115,"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":0.06080000102519989,"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/gradient-descent","display_name":"Gradient descent","score":0.6247000098228455},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.6111999750137329},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.572700023651123},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5654000043869019},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5497000217437744},{"id":"https://openalex.org/keywords/projection","display_name":"Projection (relational algebra)","score":0.5274999737739563},{"id":"https://openalex.org/keywords/adversarial-system","display_name":"Adversarial system","score":0.5157999992370605},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4668000042438507}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6481000185012817},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.6247000098228455},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.6111999750137329},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.572700023651123},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5654000043869019},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5497000217437744},{"id":"https://openalex.org/C57493831","wikidata":"https://www.wikidata.org/wiki/Q3134666","display_name":"Projection (relational algebra)","level":2,"score":0.5274999737739563},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.5157999992370605},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4668000042438507},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45170000195503235},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.41670000553131104},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4115999937057495},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.3864000141620636},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.3797000050544739},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36149999499320984},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3467999994754791},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.3463999927043915},{"id":"https://openalex.org/C2776637919","wikidata":"https://www.wikidata.org/wiki/Q624380","display_name":"Descent (aeronautics)","level":2,"score":0.34619998931884766},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3345000147819519},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.33160001039505005},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.304500013589859},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.2818000018596649},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.28049999475479126},{"id":"https://openalex.org/C55660270","wikidata":"https://www.wikidata.org/wiki/Q5164377","display_name":"Constrained optimization","level":2,"score":0.25450000166893005},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.25099998712539673}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ijcnn64981.2025.11227336","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11227336","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W1978259121","https://openalex.org/W2799194071","https://openalex.org/W2963834268","https://openalex.org/W2982756474","https://openalex.org/W3098267758","https://openalex.org/W3170092793","https://openalex.org/W3199879188","https://openalex.org/W4312394042","https://openalex.org/W4384200891","https://openalex.org/W4385573216","https://openalex.org/W4390638022","https://openalex.org/W4404764563","https://openalex.org/W4410609100","https://openalex.org/W4415797606"],"related_works":[],"abstract_inverted_index":{"As":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"are":[5,25,38],"widely":[6,162],"used,":[7],"understanding":[8],"them":[9,103],"systematically":[10],"is":[11,157,209],"key":[12],"to":[13,41,64,86,96,131,197],"improving":[14],"their":[15,19],"safety":[16],"and":[17,110,151],"realizing":[18],"full":[20],"potential.":[21],"Although":[22],"many":[23],"models":[24],"aligned":[26],"using":[27,122,173],"techniques":[28,78],"such":[29],"as":[30],"reinforcement":[31],"learning":[32],"from":[33],"human":[34],"feedback":[35],"(RLHF),":[36],"they":[37],"still":[39],"vulnerable":[40],"jailbreaking":[42,160,201],"attacks.":[43],"Some":[44],"of":[45,73,90,113,148,169],"the":[46,66,71,74,81,108,111,114,127,134,141,146,149,167,170,186],"existing":[47],"adversarial":[48],"attack":[49],"methods":[50],"search":[51],"for":[52,206],"discrete":[53,82,98],"tokens":[54,72],"that":[55,133,156,185],"may":[56,84],"jailbreak":[57],"a":[58,189],"target":[59],"model":[60],"while":[61],"others":[62],"try":[63],"optimize":[65],"continuous":[67,91],"space":[68,83],"represented":[69],"by":[70],"model\u2019s":[75],"vocabulary.":[76],"While":[77],"based":[79],"on":[80,177],"prove":[85,145],"be":[87],"inefficient,":[88],"optimization":[89,120],"token":[92],"embeddings":[93],"requires":[94],"projections":[95],"produce":[97],"tokens,":[99],"which":[100],"might":[101],"render":[102],"ineffective.":[104],"To":[105],"fully":[106],"utilize":[107],"constraints":[109],"structures":[112],"space,":[115],"we":[116],"develop":[117],"an":[118,153],"intrinsic":[119],"technique":[121,150,172,187],"exponentiated":[123],"gradient":[124],"descent":[125],"with":[126,193],"Bregman":[128],"projection":[129],"method":[130],"ensure":[132],"optimized":[135],"one-hot":[136],"encoding":[137],"always":[138],"stays":[139],"within":[140],"probability":[142],"simplex.":[143],"We":[144,165],"convergence":[147],"implement":[152],"efficient":[154],"algorithm":[155],"effective":[158],"in":[159],"several":[161],"used":[163],"LLMs.":[164],"demonstrate":[166],"efficacy":[168],"proposed":[171],"five":[174],"open-source":[175],"LLMs":[176],"four":[178],"openly":[179],"available":[180,210],"datasets.":[181],"The":[182,203],"results":[183],"show":[184],"achieves":[188],"higher":[190],"success":[191],"rate":[192],"great":[194],"efficiency":[195],"compared":[196],"three":[198],"other":[199],"state-of-the-art":[200],"techniques.":[202],"source":[204],"code":[205],"our":[207],"implementation":[208],"at:":[211],"https://github.com/sbamit/Exponentiated-Gradient-Descent-LLM-Attack":[212]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-11-14T00:00:00"}