   {"id":23917,"date":"2024-09-26T08:15:45","date_gmt":"2024-09-26T08:15:45","guid":{"rendered":"https:\/\/ideas-ncbr.pl\/?p=23917"},"modified":"2024-10-30T14:47:29","modified_gmt":"2024-10-30T14:47:29","slug":"ideas-ncbr-at-neurips-2024","status":"publish","type":"post","link":"https:\/\/ideas-ncbr.pl\/en\/ideas-ncbr-at-neurips-2024\/","title":{"rendered":"Publications at NeurIPS 2024"},"content":{"rendered":"\n<p>Only affiliations of IDEAS NCBR is noted marked below:<\/p>\n\n\n\n<p>1. <strong>Repurposing Language Models into Embedding Models: Finding the Compute-Optimal Recipe<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/arxiv.org\/abs\/2406.04165\" rel=\"nofollow noopener\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2406.04165<\/a><\/p>\n\n\n\n<p>Alicja Ziarko (IDEAS NCBR, University of Warsaw, Institute of Mathematics, Polish Academy of Sciences), Albert Q. Jiang, Bartosz Piotrowski (IDEAS NCBR), Wenda Li, Mateja Jamnik, Piotr Mi\u0142o\u015b (IDEAS NCBR, University of Warsaw, Institute of Mathematics, Polish Academy of Sciences, deepsense.ai)<\/p>\n\n\n\n<p>2. <strong>Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control<\/strong><\/p>\n\n\n\n<p>[the spotlight paper]<\/p>\n\n\n\n<p><a href=\"https:\/\/arxiv.org\/abs\/2405.16158\" rel=\"nofollow noopener\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2405.16158<\/a><\/p>\n\n\n\n<p>Michal Nauman (IDEAS NCBR, Faculty of Mathematics, Informatics, and Mechanics; University of Warsaw), Mateusz Ostaszewski, Krzysztof Jankowski, Piotr Mi\u0142o\u015b (IDEAS NCBR, Faculty of Mathematics, Informatics, and Mechanics; University of Warsaw, Polish Academy of Sciences), Marek Cygan<\/p>\n\n\n\n<p>3. <strong>Mixture of Tokens: Continuous MoE through Cross-Example Aggregation<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/arxiv.org\/abs\/2310.15961\" rel=\"nofollow noopener\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2310.15961<\/a><\/p>\n\n\n\n<p>Szymon Antoniak, Micha\u0142 Krutul (IDEAS NCBR, University of Warsaw), Maciej Pi\u00f3ro (IDEAS NCBR, Polish Academy of Sciences), Jakub Krajewski (IDEAS NCBR, University of Warsaw), Jan Ludziejewski (IDEAS NCBR, University of Warsaw), Kamil Ciebiera, Krystian Kr\u00f3l, Tomasz Odrzyg\u00f3\u017ad\u017a, Marek Cygan, Sebastian Jaszczur (IDEAS NCBR, University of Warsaw)<\/p>\n\n\n\n<p>4. <strong>Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/arxiv.org\/abs\/2310.04361\" rel=\"nofollow noopener\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2310.04361<\/a><\/p>\n\n\n\n<p>Filip Szatkowski (IDEAS NCBR, Warsaw University of Technology), Bartosz W\u00f3jcik (IDEAS NCBR, Jagiellonian University), Miko\u0142aj Pi\u00f3rczy\u0144ski, Simone Scardapane<\/p>\n\n\n\n<p>5. <strong>D-MiSo: Editing Dynamic 3D Scenes using Multi-Gaussians Soup<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/arxiv.org\/abs\/2405.14276\" rel=\"nofollow noopener\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2405.14276<\/a><\/p>\n\n\n\n<p>Joanna Waczy\u0144ska, Piotr Borycki, Joanna Kaleta, S\u0142awomir Tadeja, Przemys\u0142aw Spurek (IDEAS NCBR, Jagiellonian University)<\/p>\n\n\n\n<p>6. <strong>Task-recency bias strikes back: Adapting covariances in Exemplar-Free Class Incremental Learning<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/arxiv.org\/abs\/2409.18265\" rel=\"nofollow noopener\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2409.18265<\/a><\/p>\n\n\n\n<p>Grzegorz Rype\u015b\u0107 (Warsaw University of Technology), Sebastian Cygert (Gda\u0144sk University of Technology), Tomasz Trzci\u0144ski (Warsaw University of Technology, Tooploox), Bart\u0142omiej Twardowski (Autonomous University of Barcelona, Computer Vision Center)<\/p>\n\n\n\n<p> (all also IDEAS NCBR)<\/p>\n\n\n\n<p>7. <strong>Benchmarking Structural Inference Methods for Interacting Dynamical Systems with Synthetic Data<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/openreview.net\/forum?id=PCXvcULwiI\" rel=\"nofollow noopener\" target=\"_blank\">https:\/\/openreview.net\/forum?id=PCXvcULwiI<\/a><\/p>\n\n\n\n<p>Aoran Wang, Tsz Pan Tong, Andrzej Mizera (IDEAS NCBR), Jun Pang<\/p>\n\n\n\n<div style=\"height:100px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p>Moreover, 6 publications were accepted as workshop posters:<\/p>\n\n\n\n<p>1. <strong>Different Rates for Different Weights: Decoupled Relative Learning Rate Schedules<\/strong><\/p>\n\n\n\n<ol class=\"wp-block-list\"><\/ol>\n\n\n\n<ol class=\"wp-block-list\"><\/ol>\n\n\n\n<p>Jan Ludziejewski, Maciej Pi\u00f3ro, Kamil Ciebiera, Micha\u0142 Krutul, Jakub Krajewski<\/p>\n\n\n\n<p>2. <strong>Wait, that&#8217;s not an option: LLMs Robustness with Incorrect Multiple-Choice Options<\/strong><\/p>\n\n\n\n<p>Gracjan G\u00f3ral<\/p>\n\n\n\n<p>3. <strong>Seeing Through Their Eyes: Evaluating Visual Perspective Taking in Vision Language Models<\/strong><\/p>\n\n\n\n<p>Gracjan G\u00f3ral, Alicja Ziarko, Micha\u0142 Nauman, Maciej Wo\u0142czyk<\/p>\n\n\n\n<p>4. <strong>AI for Accelerated Materials Design &#8211; NeurIPS 2024<\/strong><\/p>\n\n\n\n<p>Grzegorz Kaszuba&nbsp;&nbsp;<\/p>\n\n\n\n<p>5. <strong>Transfer Learning in Materials Informatics: structure-property relationships through minimal but highly informative multimodal input<\/strong><\/p>\n\n\n\n<p>Grzegorz Kaszuba&nbsp;&nbsp;<\/p>\n\n\n\n<p>6. <strong>Since Faithfulness Fails: The Performance Limits of Neural Causal Discovery<\/strong><\/p>\n\n\n\n<p>Mateusz Olko, Mateusz Gajewski, \u0141ukasz Kuci\u0144ski, Piotr Mi\u0142o\u015b<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Six publications and a poster co-authored by IDEAS NCBR researchers have been accepted for the NeurIPS 2024 conference, which will be held in December in Vancouver. One of the publications received distinction as a spotlight paper. Also, another 6 publications have been accepted as workshop posters. Congratulations! <\/p>\n","protected":false},"author":27,"featured_media":23913,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":"","_links_to":"","_links_to_target":""},"categories":[86,1,30,31,79],"tags":[],"class_list":["post-23917","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news-en-en","category-bez-kategorii","category-news","category-news-en","category-news-en-2"],"acf":[],"_links":{"self":[{"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/posts\/23917","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/users\/27"}],"replies":[{"embeddable":true,"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/comments?post=23917"}],"version-history":[{"count":11,"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/posts\/23917\/revisions"}],"predecessor-version":[{"id":24306,"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/posts\/23917\/revisions\/24306"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/media\/23913"}],"wp:attachment":[{"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/media?parent=23917"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/categories?post=23917"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/ideas-ncbr.pl\/en\/wp-json\/wp\/v2\/tags?post=23917"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}