{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/residual-shuffle-exchange-networks-for-fast","title":"Residual Shuffle-Exchange Networks for Fast Processing of Long Sequences","arxiv_id":"2004.04662","date":"2020-04-06","proceeding":null,"authors":["Andis Draguns","Emīls Ozoliņš","Agris Šostaks","Matīss Apinis","Kārlis Freivalds"],"abstract":"Attention is a commonly used mechanism in sequence processing, but it is of O(n^2) complexity which prevents its application to long sequences. The recently introduced neural Shuffle-Exchange network offers a computation-efficient alternative, enabling the modelling of long-range dependencies in O(n log n) time. The model, however, is quite complex, involving a sophisticated gating mechanism derived from the Gated Recurrent Unit. In this paper, we present a simple and lightweight variant of the Shuffle-Exchange network, which is based on a residual network employing GELU and Layer Normalization. The proposed architecture not only scales to longer sequences but also converges faster and provides better accuracy. It surpasses the Shuffle-Exchange network on the LAMBADA language modelling task and achieves state-of-the-art performance on the MusicNet dataset for music transcription while being efficient in the number of parameters. We show how to combine the improved Shuffle-Exchange network with convolutional layers, establishing it as a useful building block in long sequence processing applications.","url_abs":"https://arxiv.org/abs/2004.04662v4","url_pdf":"https://arxiv.org/pdf/2004.04662v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"residual-shuffle-exchange-networks-for-fast","repo_url":"https://github.com/LUMII-Syslab/RSE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"residual-shuffle-exchange-networks-for-fast","repo_url":"https://github.com/Aroksak/RSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"lambada","task_name":"LAMBADA"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"music-transcription","task_name":"Music Transcription"}],"methods":[{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"rse","method_name":"RSE"},{"method_slug":"rsu-benes-block","method_name":"RSU Beneš Block"}],"datasets_introduced":[],"methods_introduced":[{"slug":"rse","name":"RSE","full_name":"Residual Shuffle-Exchange Network"},{"slug":"rsu-benes-block","name":"RSU Beneš Block","full_name":"Beneš Block with Residual Switch Units"}],"results":[{"leaderboard":"/sota/language-modelling-on-lambada","task":"Language Modelling","dataset":"LAMBADA","model":"Residual Shuffle-Exchange network","rank_in_archive_order":31,"of":37,"metrics":{"Accuracy":"54.34"},"uses_additional_data":false},{"leaderboard":"/sota/music-transcription-on-musicnet","task":"Music Transcription","dataset":"MusicNet","model":"Residual Shuffle-Exchange network","rank_in_archive_order":1,"of":6,"metrics":{"APS":"78.02","Number of params":"3.06M"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}