{"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/beit-bert-pre-training-of-image-transformers","title":"BEiT: BERT Pre-Training of Image Transformers","arxiv_id":"2106.08254","date":"2021-06-15","proceeding":"ICLR 2022 4","authors":["Hangbo Bao","Li Dong","Songhao Piao","Furu Wei"],"abstract":"We introduce a self-supervised vision representation model BEiT, which stands for Bidirectional Encoder representation from Image Transformers. Following BERT developed in the natural language processing area, we propose a masked image modeling task to pretrain vision Transformers. Specifically, each image has two views in our pre-training, i.e, image patches (such as 16x16 pixels), and visual tokens (i.e., discrete tokens). We first \"tokenize\" the original image into visual tokens. Then we randomly mask some image patches and fed them into the backbone Transformer. The pre-training objective is to recover the original visual tokens based on the corrupted image patches. After pre-training BEiT, we directly fine-tune the model parameters on downstream tasks by appending task layers upon the pretrained encoder. Experimental results on image classification and semantic segmentation show that our model achieves competitive results with previous pre-training methods. For example, base-size BEiT achieves 83.2% top-1 accuracy on ImageNet-1K, significantly outperforming from-scratch DeiT training (81.8%) with the same setup. Moreover, large-size BEiT obtains 86.3% only using ImageNet-1K, even outperforming ViT-L with supervised pre-training on ImageNet-22K (85.2%). The code and pretrained models are available at https://aka.ms/beit.","url_abs":"https://arxiv.org/abs/2106.08254v2","url_pdf":"https://arxiv.org/pdf/2106.08254v2.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":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/microsoft/unilm/tree/master/beit","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/E-DEEP/PapersReview","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/facebookresearch/data2vec_vision","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/facebookresearch/vissl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/huggingface/transformers","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/rwightman/pytorch-image-models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/sunsmarterjie/itpn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/woctezuma/steam-BEiT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/yyk-wew/semanticmim","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/2024-MindSpore-1/Code2/tree/main/model-1/beit","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/2024-MindSpore-1/Code6/tree/main/beit","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/leondgarse/keras_cv_attention_models/tree/main/keras_cv_attention_models/beit","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/pengzhiliang/MAE-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"beit-bert-pre-training-of-image-transformers","repo_url":"https://github.com/yangyucheng000/University/tree/main/model-2/beit","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"document-image-classification","task_name":"Document Image Classification"},{"task_slug":"document-layout-analysis","task_name":"Document Layout Analysis"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"self-supervised-image-classification","task_name":"Self-Supervised Image Classification"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"deit","method_name":"DeiT"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"feedforward-network","method_name":"Feedforward Network"},{"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":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/document-image-classification-on-rvl-cdip","task":"Document Image Classification","dataset":"RVL-CDIP","model":"BEiT-B","rank_in_archive_order":27,"of":31,"metrics":{"Accuracy":"91.09%","Parameters":"87M"},"uses_additional_data":false},{"leaderboard":"/sota/document-layout-analysis-on-publaynet-val","task":"Document Layout Analysis","dataset":"PubLayNet val","model":"BEiT-B","rank_in_archive_order":11,"of":15,"metrics":{"Figure":" 0.957","List":"0.924","Overall":"0.931","Table":"0.973","Text":"0.934","Title":"0.866"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"BEiT-L (ViT; ImageNet-22K pretrain)","rank_in_archive_order":37,"of":1060,"metrics":{"Number of params":"331M","Top 1 Accuracy":"88.60%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"BEiT-L (ViT; ImageNet 1k pretrain)","rank_in_archive_order":158,"of":1060,"metrics":{"Number of params":"86M","Top 1 Accuracy":"86.3%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-omnibenchmark","task":"Image Classification","dataset":"OmniBenchmark","model":"BeiT","rank_in_archive_order":22,"of":22,"metrics":{"Average Top-1 Accuracy":"30.1"},"uses_additional_data":false},{"leaderboard":"/sota/self-supervised-image-classification-on-1","task":"Self-Supervised Image Classification","dataset":"ImageNet (finetuned)","model":"BEiT-L (ViT)","rank_in_archive_order":14,"of":65,"metrics":{"Number of Params":"307M","Top 1 Accuracy":"86.3%"},"uses_additional_data":true},{"leaderboard":"/sota/self-supervised-image-classification-on-1","task":"Self-Supervised Image Classification","dataset":"ImageNet (finetuned)","model":"BEiT-B (ViT)","rank_in_archive_order":29,"of":65,"metrics":{"Number of Params":"86M","Top 1 Accuracy":"84.6%"},"uses_additional_data":true},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"BEiT-L (ViT+UperNet)","rank_in_archive_order":35,"of":235,"metrics":{"Validation mIoU":"57.0"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k-val","task":"Semantic Segmentation","dataset":"ADE20K val","model":"BEiT-L (ViT+UperNet, ImageNet-22k pretrain)","rank_in_archive_order":22,"of":95,"metrics":{"mIoU":"57.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2106.08254","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.08254"}},"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/rwightman/pytorch-image-models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/huggingface/transformers","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/vissl","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/E-DEEP/PapersReview","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yyk-wew/semanticmim","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/data2vec_vision","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/leondgarse/keras_cv_attention_models/tree/main/keras_cv_attention_models/beit","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yangyucheng000/University/tree/main/model-2/beit","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/woctezuma/steam-BEiT","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pengzhiliang/MAE-pytorch","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/lucidrains/DALLE-pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/microsoft/unilm/tree/master/beit","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2024-MindSpore-1/Code2/tree/main/model-1/beit","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2024-MindSpore-1/Code6/tree/main/beit","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sunsmarterjie/itpn","reach":{"status":"ok"}}],"summary":{"ran":4,"ran_fixture":2,"unverified":5},"by_repo_kind":{"official":{"samples":2,"ran":1,"repositories":1},"listed":{"samples":9,"ran":5,"repositories":2}},"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":"8d8f50f0e337e6ae","entry":"Block","repo":"microsoft/unilm","repo_kind":"official","path":"beit/modeling_pretrain.py","file_url":"https://github.com/microsoft/unilm/blob/HEAD/beit/modeling_pretrain.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8d8f50f0e337e6ae"}},{"code_sha256_prefix":"f4d723c4060a9c88","entry":"DropPath","repo":"facebookresearch/vissl","repo_kind":"listed","path":"vissl/models/trunks/beit_transformer.py","file_url":"https://github.com/facebookresearch/vissl/blob/HEAD/vissl/models/trunks/beit_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f4d723c4060a9c88"}},{"code_sha256_prefix":"b4267c3279f7dc9e","entry":"PatchEmbed","repo":"facebookresearch/data2vec_vision","repo_kind":"listed","path":"beit/modeling_pretrain.py","file_url":"https://github.com/facebookresearch/data2vec_vision/blob/HEAD/beit/modeling_pretrain.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b4267c3279f7dc9e"}},{"code_sha256_prefix":"d00d78c96a79991c","entry":"PatchEmbed","repo":"facebookresearch/vissl","repo_kind":"listed","path":"vissl/models/trunks/beit_transformer.py","file_url":"https://github.com/facebookresearch/vissl/blob/HEAD/vissl/models/trunks/beit_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d00d78c96a79991c"}},{"code_sha256_prefix":"0233d6e8a07784ae","entry":"drop_path","repo":"facebookresearch/vissl","repo_kind":"listed","path":"vissl/models/trunks/beit_transformer.py","file_url":"https://github.com/facebookresearch/vissl/blob/HEAD/vissl/models/trunks/beit_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0233d6e8a07784ae"}},{"code_sha256_prefix":"ab8464b48ee9d702","entry":"trunc_normal_","repo":"facebookresearch/vissl","repo_kind":"listed","path":"vissl/models/trunks/beit_transformer.py","file_url":"https://github.com/facebookresearch/vissl/blob/HEAD/vissl/models/trunks/beit_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ab8464b48ee9d702"}},{"code_sha256_prefix":"0b04b71ac3366634","entry":"BEiT","repo":"facebookresearch/vissl","repo_kind":"listed","path":"vissl/models/trunks/beit_transformer.py","file_url":"https://github.com/facebookresearch/vissl/blob/HEAD/vissl/models/trunks/beit_transformer.py","link_basis":"first_harvest_node","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":"0b04b71ac3366634"}},{"code_sha256_prefix":"514990276df7d1e1","entry":"Block","repo":"facebookresearch/data2vec_vision","repo_kind":"listed","path":"beit/modeling_pretrain.py","file_url":"https://github.com/facebookresearch/data2vec_vision/blob/HEAD/beit/modeling_pretrain.py","link_basis":"first_harvest_node","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":"514990276df7d1e1"}},{"code_sha256_prefix":"c8d115d51a70401e","entry":"Block","repo":"facebookresearch/vissl","repo_kind":"listed","path":"vissl/models/trunks/beit_transformer.py","file_url":"https://github.com/facebookresearch/vissl/blob/HEAD/vissl/models/trunks/beit_transformer.py","link_basis":"first_harvest_node","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":"c8d115d51a70401e"}},{"code_sha256_prefix":"c4156b300b1ba24f","entry":"VisionTransformerForMaskedImageModeling","repo":"facebookresearch/data2vec_vision","repo_kind":"listed","path":"beit/modeling_pretrain.py","file_url":"https://github.com/facebookresearch/data2vec_vision/blob/HEAD/beit/modeling_pretrain.py","link_basis":"first_harvest_node","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":"c4156b300b1ba24f"}},{"code_sha256_prefix":"40fe357c6531e542","entry":"VisionTransformerForMaskedImageModeling","repo":"microsoft/unilm","repo_kind":"official","path":"beit/modeling_pretrain.py","file_url":"https://github.com/microsoft/unilm/blob/HEAD/beit/modeling_pretrain.py","link_basis":"first_harvest_node","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":"40fe357c6531e542"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}