{"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/2103-15358","title":"Multi-Scale Vision Longformer: A New Vision Transformer for High-Resolution Image Encoding","arxiv_id":"2103.15358","date":"2021-03-29","proceeding":"ICCV 2021 10","authors":["Pengchuan Zhang","Xiyang Dai","Jianwei Yang","Bin Xiao","Lu Yuan","Lei Zhang","Jianfeng Gao"],"abstract":"This paper presents a new Vision Transformer (ViT) architecture Multi-Scale Vision Longformer, which significantly enhances the ViT of \\cite{dosovitskiy2020image} for encoding high-resolution images using two techniques. The first is the multi-scale model structure, which provides image encodings at multiple scales with manageable computational cost. The second is the attention mechanism of vision Longformer, which is a variant of Longformer \\cite{beltagy2020longformer}, originally developed for natural language processing, and achieves a linear complexity w.r.t. the number of input tokens. A comprehensive empirical study shows that the new ViT significantly outperforms several strong baselines, including the existing ViT models and their ResNet counterparts, and the Pyramid Vision Transformer from a concurrent work \\cite{wang2021pyramid}, on a range of vision tasks, including image classification, object detection, and segmentation. The models and source code are released at \\url{https://github.com/microsoft/vision-longformer}.","url_abs":"https://arxiv.org/abs/2103.15358v2","url_pdf":"https://arxiv.org/pdf/2103.15358v2.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":"2103-15358","repo_url":"https://github.com/microsoft/vision-longformer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"2103-15358","repo_url":"https://github.com/microsoft/VisionLongformerForObjectDetection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"2103-15358","repo_url":"https://github.com/microsoft/esvit","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"adamw","method_name":"AdamW"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"longformer","method_name":"Longformer"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"ViL-Medium-D","rank_in_archive_order":442,"of":1060,"metrics":{"GFLOPs":"8.7","Number of params":"39.7M","Top 1 Accuracy":"83.3%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"ViL-Base-D","rank_in_archive_order":460,"of":1060,"metrics":{"GFLOPs":"13.4","Number of params":"55.7M","Top 1 Accuracy":"83.2%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"ViL-Medium-W","rank_in_archive_order":487,"of":1060,"metrics":{"Number of params":"39.8M","Top 1 Accuracy":"82.9%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"ViL-Small","rank_in_archive_order":584,"of":1060,"metrics":{"GFLOPs":"4.86","Number of params":"24.6M","Top 1 Accuracy":"82%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"ViL-Base-W","rank_in_archive_order":600,"of":1060,"metrics":{"GFLOPs":"6.74","Number of params":"79M","Top 1 Accuracy":"81.9%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"ViL-Tiny-RPB","rank_in_archive_order":904,"of":1060,"metrics":{"GFLOPs":"1.3","Number of params":"6.7M","Top 1 Accuracy":"76.7%"},"uses_additional_data":false},{"leaderboard":"/sota/instance-segmentation-on-coco-minival","task":"Instance Segmentation","dataset":"COCO minival","model":"Mask R-CNN (ViL Base, multi-scale, 3x lr)","rank_in_archive_order":47,"of":93,"metrics":{"AP75":"49.9","mask AP":"45.7"},"uses_additional_data":false},{"leaderboard":"/sota/instance-segmentation-on-coco-minival","task":"Instance Segmentation","dataset":"COCO minival","model":"Mask R-CNN (ViL Base, 1x lr)","rank_in_archive_order":48,"of":93,"metrics":{"AP50":"67.2","AP75":"49.3","mask AP":"45.1"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"RetinaNet (ViL-Base, multi-scale, 3x)","rank_in_archive_order":124,"of":220,"metrics":{"AP75":"47.6","APL":"58.1","APM":"48","APS":"29.9","box AP":"44.7"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"RetinaNet (ViL-Base)","rank_in_archive_order":132,"of":220,"metrics":{"AP50":"65.5","AP75":"47.1","APL":"58.3","APM":"47.9","APS":"28.9","box AP":"44.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2103.15358","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2103.15358"}},"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/microsoft/VisionLongformerForObjectDetection","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/microsoft/esvit","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/microsoft/vision-longformer","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/microsoft/visionlongformer","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}}],"summary":{"ran_violates":2,"ran_draft_wrong":3,"ran_honours":1,"unverified":4},"by_repo_kind":{"official":{"samples":3,"ran":3,"repositories":1},"listed":{"samples":7,"ran":3,"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":"60fff7c3c400d7ff","entry":"default","repo":"microsoft/VisionLongformerForObjectDetection","repo_kind":"listed","path":"msvit/performer.py","file_url":"https://github.com/microsoft/VisionLongformerForObjectDetection/blob/HEAD/msvit/performer.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"60fff7c3c400d7ff"}},{"code_sha256_prefix":"aa5486a3650902d8","entry":"exists","repo":"microsoft/VisionLongformerForObjectDetection","repo_kind":"listed","path":"msvit/performer.py","file_url":"https://github.com/microsoft/VisionLongformerForObjectDetection/blob/HEAD/msvit/performer.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"aa5486a3650902d8"}},{"code_sha256_prefix":"4e291c3ece372165","entry":"find_modules","repo":"microsoft/vision-longformer","repo_kind":"official","path":"src/models/msvit.py","file_url":"https://github.com/microsoft/vision-longformer/blob/HEAD/src/models/msvit.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":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4e291c3ece372165"}},{"code_sha256_prefix":"f8ef5ca2539fcee5","entry":"get_module_device","repo":"microsoft/vision-longformer","repo_kind":"official","path":"src/models/msvit.py","file_url":"https://github.com/microsoft/vision-longformer/blob/HEAD/src/models/msvit.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":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f8ef5ca2539fcee5"}},{"code_sha256_prefix":"3444251df88e6169","entry":"init_","repo":"microsoft/vision-longformer","repo_kind":"official","path":"src/models/msvit.py","file_url":"https://github.com/microsoft/vision-longformer/blob/HEAD/src/models/msvit.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3444251df88e6169"}},{"code_sha256_prefix":"757aee54843af070","entry":"softmax_kernel","repo":"microsoft/VisionLongformerForObjectDetection","repo_kind":"listed","path":"msvit/performer.py","file_url":"https://github.com/microsoft/VisionLongformerForObjectDetection/blob/HEAD/msvit/performer.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"757aee54843af070"}},{"code_sha256_prefix":"ca8f2c57ed27ae92","entry":"mask_invalid_locations","repo":"microsoft/VisionLongformerForObjectDetection","repo_kind":"listed","path":"msvit/slidingchunk_2d.py","file_url":"https://github.com/microsoft/VisionLongformerForObjectDetection/blob/HEAD/msvit/slidingchunk_2d.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ca8f2c57ed27ae92"}},{"code_sha256_prefix":"0b417de020bf073b","entry":"resize_pos_embed_1d","repo":"microsoft/VisionLongformerForObjectDetection","repo_kind":"listed","path":"converter.py","file_url":"https://github.com/microsoft/VisionLongformerForObjectDetection/blob/HEAD/converter.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0b417de020bf073b"}},{"code_sha256_prefix":"3d16b011ef55503e","entry":"resize_pos_embed_2d","repo":"microsoft/VisionLongformerForObjectDetection","repo_kind":"listed","path":"converter.py","file_url":"https://github.com/microsoft/VisionLongformerForObjectDetection/blob/HEAD/converter.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3d16b011ef55503e"}},{"code_sha256_prefix":"56b6ba9dcdd328f4","entry":"slidingchunk_2dautograd","repo":"microsoft/VisionLongformerForObjectDetection","repo_kind":"listed","path":"msvit/slidingchunk_2d.py","file_url":"https://github.com/microsoft/VisionLongformerForObjectDetection/blob/HEAD/msvit/slidingchunk_2d.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"56b6ba9dcdd328f4"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}