{"id":39025,"date":"2026-08-21T15:28:21","date_gmt":"2026-08-21T15:28:21","guid":{"rendered":"https:\/\/lamarr-institute.org\/publication\/distillation-free-scaling-of-large-state-space-models-for-images-and-videos\/"},"modified":"2026-09-17T11:18:28","modified_gmt":"2026-09-17T11:18:28","slug":"distillation-free-scaling-of-large-state-space-models-for-images-and-videos","status":"publish","type":"publication","link":"https:\/\/lamarr-institute.org\/de\/publication\/distillation-free-scaling-of-large-state-space-models-for-images-and-videos\/","title":{"rendered":"Distillation-free Scaling of Large State-Space Models for Images and Videos"},"content":{"rendered":"<p>State-space models ({SSMs}), exemplified by S4, have introduced a novel context modeling method by integrating state-space techniques into deep learning. Despite their effectiveness, {SSMs} struggle with global context modeling due to data-independent matrices. The Mamba model addresses this with data-dependent variants enabled by the S6 selective-scan algorithm, enhancing context modeling, especially for long sequences. However, Mamba-based architectures face significant parameter scalability challenges, limiting their utility in vision applications. This paper tackles the scalability issue of large {SSMs} for image classification and action recognition without relying on additional techniques like knowledge distillation. We analyze the distinct characteristics of Mamba-based and Attention-based models, proposing a Mamba-Attention interleaved architecture that enhances scalability, robustness, and performance. We demonstrate that the stable and efficient interleaved architecture resolves the scalability issue of Mamba-based architectures and increases robustness to common corruption artifacts. Our thorough evaluation on the {ImageNet}-1K, Kinetics-400, and Something-Something-v2 benchmarks demonstrates that our approach improves the accuracy of state-of-the-art Mamba-based architectures by up to \\$\\$+1.7\\$\\$\\%.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>State-space models ({SSMs}), exemplified by S4, have introduced a novel context modeling method by integrating state-space techniques into deep learning. Despite their effectiveness, {SSMs} struggle with global context modeling due to data-independent matrices. The Mamba model addresses this with data-dependent variants enabled by the S6 selective-scan algorithm, enhancing context modeling, especially for long sequences. However, Mamba-based architectures face significant parameter scalability challenges, limiting their utility in vision applications. This paper [&hellip;]<\/p>\n","protected":false},"author":14,"featured_media":0,"template":"","meta":{"_acf_changed":false,"footnotes":""},"publication-type":[30],"class_list":["post-39025","publication","type-publication","status-publish","hentry","publication-type-article"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/39025","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication"}],"about":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/types\/publication"}],"author":[{"embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/users\/14"}],"version-history":[{"count":0,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/39025\/revisions"}],"wp:attachment":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/media?parent=39025"}],"wp:term":[{"taxonomy":"publication-type","embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication-type?post=39025"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}