{"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/unireplknet-a-universal-perception-large","title":"UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition","arxiv_id":"2311.15599","date":"2023-11-27","proceeding":null,"authors":["Xiaohan Ding","Yiyuan Zhang","Yixiao Ge","Sijie Zhao","Lin Song","Xiangyu Yue","Ying Shan"],"abstract":"Large-kernel convolutional neural networks (ConvNets) have recently received extensive research attention, but two unresolved and critical issues demand further investigation. 1) The architectures of existing large-kernel ConvNets largely follow the design principles of conventional ConvNets or transformers, while the architectural design for large-kernel ConvNets remains under-addressed. 2) As transformers have dominated multiple modalities, it remains to be investigated whether ConvNets also have a strong universal perception ability in domains beyond vision. In this paper, we contribute from two aspects. 1) We propose four architectural guidelines for designing large-kernel ConvNets, the core of which is to exploit the essential characteristics of large kernels that distinguish them from small kernels - they can see wide without going deep. Following such guidelines, our proposed large-kernel ConvNet shows leading performance in image recognition (ImageNet accuracy of 88.0%, ADE20K mIoU of 55.6%, and COCO box AP of 56.4%), demonstrating better performance and higher speed than the recent powerful competitors. 2) We discover large kernels are the key to unlocking the exceptional performance of ConvNets in domains where they were originally not proficient. With certain modality-related preprocessing approaches, the proposed model achieves state-of-the-art performance on time-series forecasting and audio recognition tasks even without modality-specific customization to the architecture. All the code and models are publicly available on GitHub and Huggingface.","url_abs":"https://arxiv.org/abs/2311.15599v2","url_pdf":"https://arxiv.org/pdf/2311.15599v2.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":"unireplknet-a-universal-perception-large","repo_url":"https://github.com/ailab-cvc/unireplknet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"unireplknet-a-universal-perception-large","repo_url":"https://github.com/Westlake-AI/openmixup","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"unireplknet-a-universal-perception-large","repo_url":"https://github.com/chenller/mmseg-extension","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series-forecasting","task_name":"Time Series Forecasting"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-XL++","rank_in_archive_order":63,"of":1060,"metrics":{"Top 1 Accuracy":"88%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-L++","rank_in_archive_order":67,"of":1060,"metrics":{"Top 1 Accuracy":"87.9%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-B++","rank_in_archive_order":91,"of":1060,"metrics":{"Top 1 Accuracy":"87.4%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-S++","rank_in_archive_order":145,"of":1060,"metrics":{"Top 1 Accuracy":"86.4%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-S","rank_in_archive_order":374,"of":1060,"metrics":{"Top 1 Accuracy":"83.9%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-T","rank_in_archive_order":449,"of":1060,"metrics":{"Top 1 Accuracy":"83.2%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-N","rank_in_archive_order":622,"of":1060,"metrics":{"Top 1 Accuracy":"81.6%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-P","rank_in_archive_order":712,"of":1060,"metrics":{"Top 1 Accuracy":"80.2%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-F","rank_in_archive_order":818,"of":1060,"metrics":{"Top 1 Accuracy":"78.6%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"UniRepLKNet-A","rank_in_archive_order":891,"of":1060,"metrics":{"Top 1 Accuracy":"77%"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-2017","task":"Object Detection","dataset":"COCO 2017","model":"UniRepLKNet-XL++","rank_in_archive_order":8,"of":24,"metrics":{"mAP":"56.4"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-2017","task":"Object Detection","dataset":"COCO 2017","model":"UniRepLKNet-L++","rank_in_archive_order":9,"of":24,"metrics":{"mAP":"55.8"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-2017","task":"Object Detection","dataset":"COCO 2017","model":"UniRepLKNet-B++","rank_in_archive_order":10,"of":24,"metrics":{"mAP":"54.8"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-2017","task":"Object Detection","dataset":"COCO 2017","model":"UniRepLKNet-S++","rank_in_archive_order":11,"of":24,"metrics":{"mAP":"54.3"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-2017","task":"Object Detection","dataset":"COCO 2017","model":"UniRepLKNet-S","rank_in_archive_order":13,"of":24,"metrics":{"mAP":"53"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-2017","task":"Object Detection","dataset":"COCO 2017","model":"UniRepLKNet-T","rank_in_archive_order":15,"of":24,"metrics":{"mAP":"51.7"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"UniRepLKNet-XL","rank_in_archive_order":46,"of":235,"metrics":{"Validation mIoU":"55.6"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"UniRepLKNet-L++","rank_in_archive_order":52,"of":235,"metrics":{"Validation mIoU":"55"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"UniRepLKNet-B++","rank_in_archive_order":72,"of":235,"metrics":{"Validation mIoU":"53.9"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"UniRepLKNet-S++","rank_in_archive_order":88,"of":235,"metrics":{"Validation mIoU":"52.7"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"UniRepLKNet-S","rank_in_archive_order":105,"of":235,"metrics":{"Validation mIoU":"51"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"UniRepLKNet-T","rank_in_archive_order":138,"of":235,"metrics":{"Validation mIoU":"49.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2311.15599","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2311.15599"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Westlake-AI/openmixup","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ailab-cvc/unireplknet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/chenller/mmseg-extension","reach":null}],"summary":{"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"9a2e756528f23042","entry":"get_bn","repo":"ailab-cvc/unireplknet","repo_kind":"official","path":"unireplknet.py","file_url":"https://github.com/ailab-cvc/unireplknet/blob/HEAD/unireplknet.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9a2e756528f23042"}},{"code_sha256_prefix":"5a70ce114f24b052","entry":"fuse_bn","repo":"ailab-cvc/unireplknet","repo_kind":"official","path":"unireplknet.py","file_url":"https://github.com/ailab-cvc/unireplknet/blob/HEAD/unireplknet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"5a70ce114f24b052"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}