{"id":1014,"date":"2022-05-10T15:47:47","date_gmt":"2022-05-10T14:47:47","guid":{"rendered":"http:\/\/lovc.cs.uni-bonn.de\/?page_id=1014"},"modified":"2023-04-12T12:54:04","modified_gmt":"2023-04-12T11:54:04","slug":"zhakshylyk-nurlanov","status":"publish","type":"page","link":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/team\/zhakshylyk-nurlanov\/","title":{"rendered":"Zhakshylyk Nurlanov"},"content":{"rendered":"<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-740\" src=\"https:\/\/lovc.cs.uni-bonn.de\/wp-content\/uploads\/2022\/05\/nurlanov_sm.jpg\" alt=\"\" width=\"120\" height=\"154\" \/><\/p>\n<p>I am a PhD student at University of Bonn and <a href=\"https:\/\/www.bosch-ai.com\/\">Bosch Center for AI<\/a> supervised by Florian Bernard and <a href=\"https:\/\/www.frank-r-schmidt.de\/\">Frank R. Schmidt<\/a>. Currently I am interested in training verifiably robust deep learning models. I did my bachelors at Moscow Institute of Physics and Technology and my masters at Technical University of Munich.<\/p>\n<p><strong>Publications:<\/strong><br \/>\n<div class=\"teachpress_pub_list\"><form name=\"tppublistform\" method=\"get\"><a name=\"tppubs\" id=\"tppubs\"><\/a><\/form><link href=\"https:\/\/lovc.cs.uni-bonn.de\/wp-content\/uploads\/teachpress\/templates\/tp_template_fb.css\" type=\"text\/css\" rel=\"stylesheet\" \/><table class=\"teachpress_publication_list\"><tr>\n                    <td colspan=\"2\">\n                        <h3 class=\"tp_h3\" id=\"tp_h3_2026\">2026<\/h3>\n                    <\/td>\n                <\/tr><tr class=\"tp_publication tp_publication_article\"><td class=\"tp_pub_image_left\" width=\"185\"><img decoding=\"async\" name=\"Jailbreaking LLMs Without Gradients or Priors: Effective and Transferable Attacks\" src=\"https:\/\/lovc.cs.uni-bonn.de\/wp-content\/uploads\/2026\/04\/nurlanov_icpr26.png\" width=\"180\" alt=\"Jailbreaking LLMs Without Gradients or Priors: Effective and Transferable Attacks\" \/><\/td><td class=\"tp_pub_info\"><div class=\"tp_pub_title\">Jailbreaking LLMs Without Gradients or Priors: Effective and Transferable Attacks<\/div><div class=\"tp_pub_author\">Z. Nurlanov, F. R. Schmidt, F. Bernard<\/div><div class=\"tp_pub_additional\"><span class=\"tp_pub_additional_in\">In: <\/span><span class=\"tp_pub_additional_journal\">International Conference on Pattern Recognition (ICPR), <\/span><span class=\"tp_pub_additional_year\">2026<\/span>.<\/div><div class=\"tp_bibtex\" id=\"tp_bibtex_519\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@article{nurlanov2026,<br \/>\r\ntitle = {Jailbreaking LLMs Without Gradients or Priors: Effective and Transferable Attacks},<br \/>\r\nauthor = {Z. Nurlanov and F. R. Schmidt and F. Bernard},<br \/>\r\nyear  = {2026},<br \/>\r\ndate = {2026-08-01},<br \/>\r\njournal = {International Conference on Pattern Recognition (ICPR)},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {article}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('519','tp_bibtex')\">Close<\/a><\/p><\/div><\/td><\/tr><tr>\n                    <td colspan=\"2\">\n                        <h3 class=\"tp_h3\" id=\"tp_h3_2024\">2024<\/h3>\n                    <\/td>\n                <\/tr><tr class=\"tp_publication tp_publication_article\"><td class=\"tp_pub_image_left\" width=\"185\"><img decoding=\"async\" name=\"Adaptive Certified Training: Towards Better Accuracy-Robustness Tradeoffs\" src=\"https:\/\/lovc.cs.uni-bonn.de\/wp-content\/uploads\/2023\/06\/nurlanov23.png\" width=\"180\" alt=\"Adaptive Certified Training: Towards Better Accuracy-Robustness Tradeoffs\" \/><\/td><td class=\"tp_pub_info\"><div class=\"tp_pub_title\"><a class=\"tp_title_link\" href=\"https:\/\/arxiv.org\/abs\/2307.13078\" title=\"https:\/\/arxiv.org\/abs\/2307.13078\" target=\"blank\">Adaptive Certified Training: Towards Better Accuracy-Robustness Tradeoffs<\/a><\/div><div class=\"tp_pub_author\">Z. Nurlanov, F. R. Schmidt, F. Bernard<\/div><div class=\"tp_pub_additional\"><span class=\"tp_pub_additional_in\">In: <\/span><span class=\"tp_pub_additional_journal\">European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), <\/span><span class=\"tp_pub_additional_year\">2024<\/span>.<\/div><a class=\"tp_bottom_urls\" target=\"_blank\" href=\"https:\/\/arxiv.org\/abs\/2307.13078\">[pdf] <\/a><a class=\"tp_bottom_urls\" target=\"_blank\" href=\"https:\/\/github.com\/boschresearch\/adaptive_robust_training\">[code] <\/a><div class=\"tp_bibtex\" id=\"tp_bibtex_488\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@article{nurlanov24,<br \/>\r\ntitle = {Adaptive Certified Training: Towards Better Accuracy-Robustness Tradeoffs},<br \/>\r\nauthor = {Z. Nurlanov and F. R. Schmidt and F. Bernard},<br \/>\r\nurl = {https:\/\/arxiv.org\/abs\/2307.13078},<br \/>\r\nyear  = {2024},<br \/>\r\ndate = {2024-09-09},<br \/>\r\nurldate = {2024-09-09},<br \/>\r\njournal = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD)},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {article}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('488','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_488\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"ai ai-arxiv\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/arxiv.org\/abs\/2307.13078\" title=\"https:\/\/arxiv.org\/abs\/2307.13078\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2307.13078<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('488','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><tr>\n                    <td colspan=\"2\">\n                        <h3 class=\"tp_h3\" id=\"tp_h3_2023\">2023<\/h3>\n                    <\/td>\n                <\/tr><tr class=\"tp_publication tp_publication_article\"><td class=\"tp_pub_image_left\" width=\"185\"><img decoding=\"async\" name=\"Universe Points Representation Learning for Partial Multi-Graph Matching\" src=\"https:\/\/lovc.cs.uni-bonn.de\/wp-content\/uploads\/2023\/02\/MGM-URL2.jpg\" width=\"180\" alt=\"Universe Points Representation Learning for Partial Multi-Graph Matching\" \/><\/td><td class=\"tp_pub_info\"><div class=\"tp_pub_title\"><a class=\"tp_title_link\" href=\"https:\/\/arxiv.org\/abs\/2212.00780\" title=\"https:\/\/arxiv.org\/abs\/2212.00780\" target=\"blank\">Universe Points Representation Learning for Partial Multi-Graph Matching<\/a><\/div><div class=\"tp_pub_author\">Z. Nurlanov, F. R. Schmidt, F. Bernard <\/div><div class=\"tp_pub_additional\"><span class=\"tp_pub_additional_in\">In: <\/span><span class=\"tp_pub_additional_journal\">AAAI Conference on Artificial Intelligence, <\/span><span class=\"tp_pub_additional_year\">2023<\/span>.<\/div><a class=\"tp_bottom_urls\" target=\"_blank\" href=\"https:\/\/arxiv.org\/abs\/2212.00780\">[pdf] <\/a><br><span class=\"tp_pub_menu\" style=\"color:red\">selected as oral <\/span><div class=\"tp_bibtex\" id=\"tp_bibtex_480\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@article{nurlanov2023,<br \/>\r\ntitle = {Universe Points Representation Learning for Partial Multi-Graph Matching},<br \/>\r\nauthor = {Z. Nurlanov and F. R. Schmidt and F. Bernard },<br \/>\r\nurl = {https:\/\/arxiv.org\/abs\/2212.00780},<br \/>\r\nyear  = {2023},<br \/>\r\ndate = {2023-02-07},<br \/>\r\nurldate = {2023-02-07},<br \/>\r\njournal = {AAAI Conference on Artificial Intelligence},<br \/>\r\nabstract = {Many challenges from natural world can be formulated as a graph matching problem. Previous deep learning-based methods mainly consider a full two-graph matching setting. In this work, we study the more general partial matching problem with multi-graph cycle consistency guarantees. Building on a recent progress in deep learning on graphs, we propose a novel data-driven method (URL) for partial multi-graph matching, which uses an object-to-universe formulation and learns latent representations of abstract universe points. The proposed approach advances the state of the art in semantic keypoint matching problem, evaluated on Pascal VOC, CUB, and Willow datasets. Moreover, the set of controlled experiments on a synthetic graph matching dataset demonstrates the scalability of our method to graphs with large number of nodes and its robustness to high partiality.},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {article}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('480','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_abstract\" id=\"tp_abstract_480\" style=\"display:none;\"><div class=\"tp_abstract_entry\">Many challenges from natural world can be formulated as a graph matching problem. Previous deep learning-based methods mainly consider a full two-graph matching setting. In this work, we study the more general partial matching problem with multi-graph cycle consistency guarantees. Building on a recent progress in deep learning on graphs, we propose a novel data-driven method (URL) for partial multi-graph matching, which uses an object-to-universe formulation and learns latent representations of abstract universe points. The proposed approach advances the state of the art in semantic keypoint matching problem, evaluated on Pascal VOC, CUB, and Willow datasets. Moreover, the set of controlled experiments on a synthetic graph matching dataset demonstrates the scalability of our method to graphs with large number of nodes and its robustness to high partiality.<\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('480','tp_abstract')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_480\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"ai ai-arxiv\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/arxiv.org\/abs\/2212.00780\" title=\"https:\/\/arxiv.org\/abs\/2212.00780\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2212.00780<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('480','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><tr>\n                    <td colspan=\"2\">\n                        <h3 class=\"tp_h3\" id=\"tp_h3_2022\">2022<\/h3>\n                    <\/td>\n                <\/tr><tr class=\"tp_publication tp_publication_article\"><td class=\"tp_pub_image_left\" width=\"205\"><img decoding=\"async\" width=180 src=\"https:\/\/lovc.cs.uni-bonn.de\/wp-content\/uploads\/images\/nurlanov2022.png\"\/><\/td><td class=\"tp_pub_info\"><div class=\"tp_pub_title\"><a class=\"tp_title_link\" href=\"https:\/\/arxiv.org\/abs\/2206.09596\" title=\"https:\/\/arxiv.org\/abs\/2206.09596\" target=\"blank\">Efficient and Flexible Sublabel-Accurate Energy Minimization<\/a><\/div><div class=\"tp_pub_author\">Z. Nurlanov, D. Cremers, F. Bernard<\/div><div class=\"tp_pub_additional\"><span class=\"tp_pub_additional_in\">In: <\/span><span class=\"tp_pub_additional_journal\">International Conference on Pattern Recognition (ICPR), <\/span><span class=\"tp_pub_additional_year\">2022<\/span>.<\/div><a class=\"tp_bottom_urls\" target=\"_blank\" href=\"https:\/\/arxiv.org\/abs\/2206.09596\">[pdf] <\/a><a class=\"tp_bottom_urls\" target=\"_blank\" href=\"https:\/\/github.com\/nurlanov-zh\/sublabel-accurate-alpha-expansion\">[code] <\/a><div class=\"tp_bibtex\" id=\"tp_bibtex_475\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@article{nurlanov2022b,<br \/>\r\ntitle = {Efficient and Flexible Sublabel-Accurate Energy Minimization},<br \/>\r\nauthor = {Z. Nurlanov and D. Cremers and F. Bernard},<br \/>\r\nurl = {https:\/\/arxiv.org\/abs\/2206.09596},<br \/>\r\nyear  = {2022},<br \/>\r\ndate = {2022-05-17},<br \/>\r\nurldate = {2022-05-17},<br \/>\r\njournal = {International Conference on Pattern Recognition (ICPR)},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {article}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('475','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_475\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"ai ai-arxiv\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/arxiv.org\/abs\/2206.09596\" title=\"https:\/\/arxiv.org\/abs\/2206.09596\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2206.09596<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('475','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><\/table><\/div><\/p>\n","protected":false},"excerpt":{"rendered":"<p>I am a PhD student at University of Bonn and Bosch Center for AI supervised by Florian Bernard and Frank R. Schmidt. Currently I am interested in training verifiably robust deep learning models. I did my bachelors at Moscow Institute of Physics and Technology and my masters at Technical University of Munich. Publications:<\/p>\n","protected":false},"author":6,"featured_media":0,"parent":697,"menu_order":1,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_acf_changed":false,"footnotes":""},"class_list":["post-1014","page","type-page","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/pages\/1014","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/comments?post=1014"}],"version-history":[{"count":11,"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/pages\/1014\/revisions"}],"predecessor-version":[{"id":1224,"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/pages\/1014\/revisions\/1224"}],"up":[{"embeddable":true,"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/pages\/697"}],"wp:attachment":[{"href":"https:\/\/lovc.cs.uni-bonn.de\/index.php\/wp-json\/wp\/v2\/media?parent=1014"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}