{"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/reversible-column-networks","title":"Reversible Column Networks","arxiv_id":"2212.11696","date":"2022-12-22","proceeding":null,"authors":["Yuxuan Cai","Yizhuang Zhou","Qi Han","Jianjian Sun","Xiangwen Kong","Jun Li","Xiangyu Zhang"],"abstract":"We propose a new neural network design paradigm Reversible Column Network (RevCol). The main body of RevCol is composed of multiple copies of subnetworks, named columns respectively, between which multi-level reversible connections are employed. Such architectural scheme attributes RevCol very different behavior from conventional networks: during forward propagation, features in RevCol are learned to be gradually disentangled when passing through each column, whose total information is maintained rather than compressed or discarded as other network does. Our experiments suggest that CNN-style RevCol models can achieve very competitive performances on multiple computer vision tasks such as image classification, object detection and semantic segmentation, especially with large parameter budget and large dataset. For example, after ImageNet-22K pre-training, RevCol-XL obtains 88.2% ImageNet-1K accuracy. Given more pre-training data, our largest model RevCol-H reaches 90.0% on ImageNet-1K, 63.8% APbox on COCO detection minival set, 61.0% mIoU on ADE20k segmentation. To our knowledge, it is the best COCO detection and ADE20k segmentation result among pure (static) CNN models. Moreover, as a general macro architecture fashion, RevCol can also be introduced into transformers or other neural networks, which is demonstrated to improve the performances in both computer vision and NLP tasks. We release code and models at https://github.com/megvii-research/RevCol","url_abs":"https://arxiv.org/abs/2212.11696v3","url_pdf":"https://arxiv.org/pdf/2212.11696v3.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":"reversible-column-networks","repo_url":"https://github.com/megvii-research/revcol","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"RevCol-H","rank_in_archive_order":11,"of":1060,"metrics":{"Number of params":"2158M","Top 1 Accuracy":"90.0%"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"RevCol-H(DINO)","rank_in_archive_order":12,"of":220,"metrics":{"box AP":"63.8"},"uses_additional_data":true},{"leaderboard":"/sota/object-detection-on-coco","task":"Object Detection","dataset":"COCO test-dev","model":"RevCol-H(DINO)","rank_in_archive_order":13,"of":225,"metrics":{"box mAP":"63.8"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"RevCol-H (Mask2Former)","rank_in_archive_order":10,"of":235,"metrics":{"Params (M)":"2439","Validation mIoU":"61.0"},"uses_additional_data":true}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2212.11696","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2212.11696"}},"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":"deterministic:regex_extraction","url":"https://github.com/megvii-research/RevCol","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/megvii-research/revcol","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":8,"unverified":3},"by_repo_kind":{"official":{"samples":11,"ran":8,"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":"62fc75450a52fc56","entry":"Classifier","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"62fc75450a52fc56"}},{"code_sha256_prefix":"008b0e076048797d","entry":"ConvNextBlock","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"008b0e076048797d"}},{"code_sha256_prefix":"0aa0f3056567a82f","entry":"Decoder","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0aa0f3056567a82f"}},{"code_sha256_prefix":"7bc303c0be07360e","entry":"Fusion","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7bc303c0be07360e"}},{"code_sha256_prefix":"3d31e56894e46115","entry":"LayerNorm","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3d31e56894e46115"}},{"code_sha256_prefix":"1939d987f30fd46b","entry":"ReverseFunction","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1939d987f30fd46b"}},{"code_sha256_prefix":"9da4f6ea726fdfb1","entry":"SimDecoder","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9da4f6ea726fdfb1"}},{"code_sha256_prefix":"a214203738550dde","entry":"UpSampleConvnext","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a214203738550dde"}},{"code_sha256_prefix":"c57688066cd3ac52","entry":"FullNet","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.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":"c57688066cd3ac52"}},{"code_sha256_prefix":"d6bc6096b81fd178","entry":"Level","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.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":"d6bc6096b81fd178"}},{"code_sha256_prefix":"6d63037ed95088f7","entry":"SubNet","repo":"megvii-research/revcol","repo_kind":"official","path":"models/revcol.py","file_url":"https://github.com/megvii-research/revcol/blob/HEAD/models/revcol.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":"6d63037ed95088f7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}