Meta tags:
description=Johns Hopkins University;
author=Chaowei Xiao;

Headings (most frequently used words):

recent, chaowei, xiao, award, invited, talks, news, selected, publications, full, list, current, phd, students,

Text of the page (most frequently used words):
and (55), 2024 (42), #chaowei (36), xiao (36), the (28), for (27), liu (24), pdf (23), models (22), 2023 (22), security (21), 2022 (19), neurips (18), accepted (16), language (15), talk (15), papers (15), iclr (13), xiaogeng (12), paper (12), invited (12), with (11), learning (11), adversarial (11), model (11), 2025 (11), our (11), attacks (10), machine (10), wang (10), anandkumar (10), llm (10), large (10), safety (10), workshop (10), from (9), chen (9), anima (9), agents (9), one (9), phd (8), will (8), two (8), song (7), based (7), nie (7), agent (7), jailbreak (7), got (7), award (7), zhang (6), madison (6), cvpr (6), dawn (6), zhu (6), privacy (6), weili (6), research (6), university (6), via (5), ning (5), zhao (5), jhu (5), vision (5), international (5), robustness (5), guo (5), backdoor (5), llms (5), retrieval (5), foundation (5), acl (5), secure (5), usenix (5), congratulations (5), group (5), zhengyue (4), conference (4), computer (4), physical (4), world (4), intelligence (4), mingyan (4), code (4), jiongxiao (4), diffusion (4), defense (4), time (4), tuning (4), naacl (4), muhao (4), somesh (4), jha (4), patrick (4), mcdaniel (4), trustworthy (4), emnlp (4), nature (4), alignment (4), era (4), acm (4), three (4), this (4), track (4), have (4), hao (3), yingzi (3), students (3), ieee (3), 2018 (3), symposium (3), yang (3), multi (3), driving (3), icml (3), huang (3), training (3), red (3), teaming (3), wei (3), linxi (3), fan (3), zhiding (3), study (3), eccv (3), sun (3), prompt (3), lifelong (3), shengchao (3), text (3), against (3), injection (3), attack (3), zhiyuan (3), exploring (3), systems (3), list (3), giving (3), has (3), won (3), gordon (3), bell (3), prize (3), are (3), first (3), recently (3), stay (3), tuned (3), tutorial (3), benchmark (3), generative (3), aims (3), competition (3), interested (3), advised (2), luo (2), xiang (2), transfer (2), current (2), robust (2), warren (2), jun (2), yan (2), generating (2), examples (2), networks (2), oakland (2), 2021 (2), yulong (2), cao (2), jin (2), densepure (2), understanding (2), manli (2), shu (2), tom (2), goldstein (2), instruction (2), zhaorun (2), memory (2), poisoning (2), fingerprinting (2), yixuan (2), hallucination (2), detection (2), free (2), ping (2), zihan (2), mohammad (2), shoeybi (2), yuke (2), bryan (2), catanzaro (2), zero (2), shot (2), edward (2), johns (2), tmlr (2), jiang (2), open (2), embodied (2), jiachen (2), marco (2), pavone (2), multimodal (2), test (2), chengpeng (2), ling (2), hongyu (2), drug (2), editing (2), domain (2), jiarui (2), jian (2), tang (2), molecule (2), design (2), framework (2), ai4science (2), fangzhou (2), system (2), level (2), guardrail (2), mitigating (2), huan (2), jiazhao (2), 2026 (2), safe (2), liang (2), prompts (2), new (2), platform (2), autodan (2), google (2), scholar (2), selected (2), publications (2), aaai (2), all (2), survey (2), been (2), special (2), covid (2), how (2), use (2), state (2), more (2), work (2), impact (2), argonne (2), national (2), laboratory (2), topic (2), propose (2), moleculestm (2), representation (2), tdw (2), currently (2), regular (2), senior (2), area (2), chair (2), interns (2), lab (2), thank (2), you (2), package (2), recent (2), best (2), stanford (2), nvidia (2), schmidt (2), sciences (2), multiple (2), cybersecurity (2), robotics (2), domains (2), assistant (2), professor (2), home (2), designed, bootstrap, washu, eddy, zheng, uga, pattern, recognition, kevin, eykholt, tadayoshi, kohno, atul, prakash, amir, rahmati, earlence, fernandes, ivan, evtimov, joint, artificial, ijcai, representations, spatially, transformed, ningfei, dawei, fang, ruigangyang, alfred, invisible, both, camera, lidar, sensor, fusion, perception, autonomous, under, zhongzhu, kun, certification, towards, brandon, yujia, arash, vahdat, purification, jonas, geiping, threats, exploitability, zhen, agentpoison, knowledge, base, jiashu, fei, mingyu, derek, pang, koh, instructional, spotlight, xuefeng, haloscope, harnessing, unlabeled, generations, watermark, can, watermarks, used, detect, infringement, zhuolin, vijay, anand, korthikanti, shiyi, lan, ming, vilm, augmented, visual, few, image, captioning, boxin, peng, lawrence, mcafee, dong, oleksii, kuchaiev, shall, pretrain, autoregressive, comprehensive, tlmr, shikun, prismer, task, experts, guanzhi, yuqi, xie, yunfan, ajay, mandlekar, voyager, ended, dolphins, generalization, adarsh, kumarappan, tiwari, peiyang, robert, joseph, george, leanagent, formal, theorem, proving, yijin, chatdrug, chatgpt, powered, conversational, using, feedback, zhuoran, qiao, animashree, modal, structure, yanjing, zhuoxinran, anthony, gitter, yutao, arvind, ramanathan, proteindt, guided, protein, bio, math, ethan, cecchetti, indirect, information, flow, control, perspective, piguard, overdefense, weidi, shenghong, dai, suman, banerjee, agrail, effective, adaptive, hung, chun, chiu, dianqi, drift, dynamic, rule, isolation, securing, solutions, yiquan, xiangyu, junjie, fine, enhanced, armor, aligning, meticulous, reasoning, liwei, yejin, choi, mid, internalizing, foundational, capability, mitigation, shuning, zach, cameron, not, listen, concerns, real, xun, haibo, tong, chengquan, yuzhou, jiawei, mintong, kang, chejian, qichang, tianneng, shi, sanmi, koyejo, percy, wenbo, decodingtrust, dtap, controllable, interactive, nan, stealthy, aligned, peiran, suh, yevgeniy, vorobeychik, zhuoqing, mao, turbo, strategy, self, exploration, evaluation, represents, equal, contribution, full, beyond, 1st, practical, deep, wild, thanks, collaborators, discuss, challenge, opportunity, computing, virtual, seminar, series, challenges, opportunities, corl, oral, presentations, explore, train, trajectory, prediction, robusttraj, team, explain, why, improve, which, leverages, pretrained, classifier, provide, art, certified, attemp, discovery, release, line, give, trustworthiness, yiming, intern, asu, copyright, generation, zhuofeng, attention, method, defend, nlp, threat, injecting, ads, that, views, property, align, into, same, space, preda, personalized, federated, five, main, finding, fingerprint, these, exciting, fields, serving, vlm, them, colm, rlfh, four, winter, break, dec, members, enjoy, some, well, deserved, vacation, their, families, loved, ones, received, funding, donations, amazon, apple, fall, ovcr, authors, news, combating, issues, nsf, network, trillion, parameter, consortium, tpc, keynote, ccs, analysis, sfu, 5th, lead, discussion, session, talks, 2014, mobicom, embedded, wireless, ewsn, hpc, elsevier, top, scientists, student, fellowship, final, ai2050, early, career, fellow, distinguished, jailbreakv, second, safebench, plans, recruit, sponsored, philanthropy, general, vla, looking, postdoc, who, experience, software, engineering, obtained, michigan, ann, arbor, bachelor, degree, tsinghua, before, joining, spent, wonderful, years, wisconsin, arizona, build, additionally, become, principled, methods, diverse, application, including, visiting, researcher, email, chaoweixiao, edu, about, hopkins,

Text of the page (random words):
chaowei xiao johns hopkins university home home current about publications chaowei xiao email chaoweixiao jhu edu i am chaowei xiao currently visiting stanford university i am an assistant professor at jhu and a researcher at nvidia research my research aims to build safe and secure ai and agents additionally i have recently become interested in exploring principled methods for diverse application domains including ai4science embodied agents and computer use agents i obtained my ph d from the university of michigan ann arbor and my bachelor s degree from tsinghua university before joining jhu i spent three wonderful years at university of wisconsin madison and arizona state university as an assistant professor google scholar i m looking for multiple postdoc who has experience in cybersecurity software engineering security rl or the robotics domain our group plans to recruit multiple phd students sponsored by schmidt sciences and open philanthropy i am interested in students in general ai cybersecurity or robotics domains interested in vla world models award 2025 jailbreakv won second prize of safebench competition 2024 usenix security distinguished paper award 2024 schmidt sciences ai2050 early career fellow 2024 acm gordon bell final list 2024 my phd student xiaogeng won nvidia fellowship on security track 2024 selected in stanford elsevier top 2 scientists list 2024 2023 acm gordon bell special prize for hpc based covid 19 research 2023 impact award from argonne national laboratory 2021 international conference on embedded wireless systems and networks ewsn best paper award 2014 mobicom best paper award recent invited talks 12 2025 i will be at neurips and lead a discussion session at alignment workshop sd on how to secure ai agents 6 2025 invited talk at the 5th workshop of adversarial machine learning on computer vision foundation models x at cvpr 5 2025 invited talk at secure generative ai agents workshop at ieee symposium on security and privacy 4 2025 invited talk at international symposium on trustworthy foundation models 12 2024 invited talk at sfu neurips 12 2024 invited talk at llm and agent safety competition neurips 10 2024 keynote at ccs workshop on large ai systems and models with privacy and safety analysis 10 2024 invited talk at trillion parameter consortium tpc on llm safety and security 10 2024 invited talk at nsf workshop on large language models for network security 06 2024 invited talk at cvpr workshop of adversarial machine learning on computer vision robustness of foundation models 06 2024 talk at naacl tutorial on combating security and privacy issues in the era of large language models 05 2024 invited talk at iclr secure and trustworthy large language models 12 2023 invited talk at neurips tdw workshop recent news package package 3 2026 senior area chair of acl emnlp and neurips 1 2025 we have 9 papers at iclr and 3 papers at acl on model safety and security congratulations to all authors 12 2024 we got the fall research competition award at uw madison thank you uw madison and ovcr 11 2024 our group recently received funding and donations thank you amazon and apple 11 2024 our lab will have a winter break this dec lab members will enjoy some well deserved vacation time with their families and loved ones 09 2024 we have four papers at neurips regular track 09 2024 our study on safety of rlfh alignment is accepted to s p oakland 2025 07 2024 multimodal jailbreak benchmark is accepted to colm it is from the interns in my group 07 2024 4 4 papers are accepted to eccv on the topic of trustworthy vlm and driving two of them are from interns in my group 06 2024 senior area chair for neurips benchmark track 06 2024 i am currently serving as ac for the neurips regular track 05 2024 our jailbreak paper is accepted to usenix security congratulations zhiyuan 03 2024 five papers at naacl on llm security 4 main and 1 finding two on the backdoor attack one on backdoor defense one on jailbreak attacks and one on model fingerprint stay tuned on these exciting fields 03 2024 preda for personalized federated learning is accepted at cvpr 2024 01 2024 three papers at iclr 01 2024 two papers at tmlr 12 2023 invited talk at neurips tdw workshop 10 2023 our paper moleculestm has been accepted to nature machine intelligence moleculestm aims to align the nature language and molecule representation into the same representation space more 10 2023 three papers at emnlp and one paper at neurips for our neurips paper we study a new threat of the instruction tuning of llms by injecting the ads this is the first work that views the llms as the generative model and aims to attack the generative property of llms 10 2023 our tutorial on security and privacy in the era of large language models is accepted to naacl 05 2023 one paper at acl congratulations to zhuofeng and jiazhao we propose an attention based method to defend against nlp backdoor attacks 04 2023 two papers at icml congratulations to jiachen and zhiyuan we propose the first benchmark for code copyright of code generation models 02 2023 two papers at cvpr congratulations to yiming and xiaogeng xiaogeng is an intern from my group at asu 02 2023 i will give a tutorial at cvpr 2023 on the topic of trustworthiness in the era of foundation models stay tuned 01 2023 impact award from argonne national laboratory 01 2023 one paper got accepted to usenix security 2023 1 2023 three papers are accepted to iclr 2023 a we explain why and how to use diffusion model to improve adversarial robustness and design densepure which leverages the pretrained diffusion model and classifier to provide the state of the art certified robustness b this is our first attemp on retrieval based framework and ai for drug discovery we will recently release more work in this research line stay tuned 12 2022 our team won the acm gordon bell special prize for covid 19 research 09 2022 one papers got accepted to usenix security 2023 09 2022 two papers got accepted to neurips 2022 09 2022 our paper robusttraj has been accepted to corl for oral presentations we explore to train a robust trajectory prediction model against adversarial attacks 08 2022 i will be giving a talk in virtual seminar series on challenges and opportunities for security privacy in machine learning 07 2022 one survey paper to discuss the challenge and opportunity of machine learning security got accepted to acm computing survey 2022 07 2022 two papers got accepted to eccv 2022 05 2022 two papers got accepted to icml 2022 thanks for all of my collaborators 3 2022 i will be giving a talk in aaai 2022 1st international workshop on practical deep learning in the wild 3 2022 i will be giving a talk in aaai 2022 workshop on adversarial machine learning and beyond 2 2022 one paper is accepted to iclr selected publications full list represents equal contribution google scholar model agent safety and robustness evaluation red teaming autodan turbo a lifelong agent for strategy self exploration to jailbreak llms xiaogeng liu peiran li g edward suh yevgeniy vorobeychik zhuoqing mao somesh jha patrick mcdaniel huan sun bo li chaowei xiao iclr 2024 autodan generating stealthy jailbreak prompts on aligned large language models xiaogeng liu nan xu muhao chen chaowei xiao iclr 2024 decodingtrust agent platform dtap a controllable and interactive red teaming platform for ai agents zhaorun chen xun liu haibo tong chengquan guo yuzhou nie jiawei zhang mintong kang chejian xu qichang liu xiaogeng liu tianneng shi chaowei xiao sanmi koyejo percy liang wenbo guo dawn song bo li a new era in llm security exploring security concerns in real world llm based systems fangzhou wu ning zhang somesh jha patrick mcdaniel chaowei xiao pdf do not listen to me understanding and exploring jailbreak prompts of large language models zhiyuan yu xiaogeng liu shuning liang zach cameron chaowei xiao ning zhang usenix security 2024 pdf safety alignment mitigation safety mid training internalizing llm safety as a foundational capability zhengyue zhao liwei jiang yejin choi chaowei xiao pdf armor aligning secure and safe large language models via meticulous reasoning zhengyue zhao yingzi ma somesh jha marco pavone patrick mcdaniel chaowei xiao iclr 2026 pdf mitigating fine tuning jailbreak attack with backdoor enhanced alignment jiongxiao wang jiazhao li yiquan li xiangyu qi muhao chen junjie hu yixuan li bo li chaowei xiao neurips 2024 agent security via system level solutions drift dynamic rule based defense with injection isolation for securing llm agents hao li xiaogeng liu hung chun chiu dianqi li ning zhang chaowei xiao neurips 2025 pdf agrail a lifelong agent guardrail with effective and adaptive safety detection weidi luo shenghong dai xiaogeng liu suman banerjee huan sun muhao chen chaowei xiao acl 2025 pdf piguard prompt injection guardrail via mitigating overdefense for free hao li xiaogeng liu ning zhang chaowei xiao acl 2025 pdf system level defense against indirect prompt injection attacks an information flow control perspective fangzhou wu ethan cecchetti chaowei xiao pdf ai4science bio and math proteindt a text guided protein design framework shengchao liu yanjing li zhuoxinran li anthony gitter yutao zhu jiarui lu zhao xu weili nie arvind ramanathan chaowei xiao jian tang hongyu guo anima anandkumar nature machine intelligence 2025 pdf multi modal molecule structure text model for text based retrieval and editing shengchao liu weili nie chengpeng wang jiarui lu zhuoran qiao ling liu jian tang chaowei xiao animashree anandkumar nature machine intelligence pdf chatdrug chatgpt powered conversational drug editing using retrieval and domain feedback shengchao liu jiongxiao wang yijin yang chengpeng wang ling liu hongyu guo chaowei xiao iclr 2024 pdf leanagent lifelong learning for formal theorem proving adarsh kumarappan mo tiwari peiyang song robert joseph george chaowei xiao anima anandkumar iclr 2024 foundation models agents test time training test time prompt tuning for zero shot generalization in vision language models manli shu weili nie de an huang zhiding yu tom goldstein anima anandkumar chaowei xiao neurips 2022 pdf dolphins multimodal language model for driving yingzi ma yulong cao jiachen sun marco pavone chaowei xiao eccv 2024 voyager an open ended embodied agent with large language models guanzhi wang yuqi xie yunfan jiang ajay mandlekar chaowei xiao yuke zhu linxi fan anima anandkumar tmlr 2024 prismer a vision language model with multi task experts shikun liu linxi fan edward johns zhiding yu chaowei xiao anima anandkumar tlmr 2024 pdf shall we pretrain autoregressive language models with retrieval a comprehensive study boxin wang wei ping peng xu lawrence mcafee zihan liu mohammad shoeybi yi dong oleksii kuchaiev bo li chaowei xiao anima anandkumar bryan catanzaro emnlp 2023 pdf re vilm retrieval augmented visual language model for zero and few shot image captioning zhuolin yang wei ping zihan liu vijay anand korthikanti weili nie de an huang linxi fan zhiding yu shiyi lan bo li mohammad shoeybi ming yu liu yuke zhu bryan catanzaro chaowei xiao anima anandkumar emnlp 2023 pdf trustworthy llms watermark can watermarks be used to detect llm ip infringement for free zhengyue zhao xiaogeng liu somesh jha patrick mcdaniel bo li chaowei xiao iclr 2024 hallucination haloscope harnessing unlabeled llm generations for hallucination detection xuefeng du chaowei xiao yixuan li neurips 2024 spotlight fingerprinting instructional fingerprinting of large language models jiashu xu fei wang mingyu derek ma pang wei koh chaowei xiao muhao chen naacl 2024 pdf memory poisoning agentpoison red teaming llm agents via memory or knowledge base backdoor poisoning zhaorun chen zhen xiang chaowei xiao dawn song bo li neurips 2024 pdf training time threats on the exploitability of instruction tuning manli shu jiongxiao wang chen zhu jonas geiping chaowei xiao tom goldstein neurips 2023 pdf adversarial machine learning defense diffusion models for adversarial purification weili nie brandon guo yujia huang chaowei xiao arash vahdat anima anandkumar icml 2022 pdf code certification densepure understanding diffusion models towards adversarial robustness chaowei xiao zhongzhu chen kun jin jiongxiao wang weili nie mingyan liu anima anandkumar bo li dawn song iclr 2023 pdf code physical attacks invisible for both camera and lidar security of multi sensor fusion based perception in autonomous driving under physical world attacks yulong cao ningfei wang chaowei xiao dawei yang jin fang ruigangyang qi alfred chen mingyan liu bo li ieee symposium on security and privacy oakland 2021 attacks spatially transformed adversarial examples chaowei xiao jun yan zhu bo li warren he mingyan liu and dawn song in international conference on learning representations iclr 2018 pdf attacks generating adversarial examples with adversarial networks chaowei xiao bo li jun yan zhu warren he mingyan liu and dawn song in international joint conference on artificial intelligence ijcai 2018 pdf physical attacks robust physical world attacks on machine learning models kevin eykholt ivan evtimov earlence fernandes bo li amir rahmati chaowei xiao atul prakash tadayoshi kohno and dawn song in ieee conference on computer vision and pattern recognition cvpr 2018 pdf current phd students xiaogeng liu phd jhu transfer from uw madison zhengyue zhao phd jhu transfer from uw madison yingzi ma phd uw madison eddy luo co advised with zheng xiang phd uga hao li co advised with ning zhang phd washu designed via bootstrap