{"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/vl-bert-pre-training-of-generic-visual","title":"VL-BERT: Pre-training of Generic Visual-Linguistic Representations","arxiv_id":"1908.08530","date":"2019-08-22","proceeding":"ICLR 2020 1","authors":["Weijie Su","Xizhou Zhu","Yue Cao","Bin Li","Lewei Lu","Furu Wei","Jifeng Dai"],"abstract":"We introduce a new pre-trainable generic representation for visual-linguistic tasks, called Visual-Linguistic BERT (VL-BERT for short). VL-BERT adopts the simple yet powerful Transformer model as the backbone, and extends it to take both visual and linguistic embedded features as input. In it, each element of the input is either of a word from the input sentence, or a region-of-interest (RoI) from the input image. It is designed to fit for most of the visual-linguistic downstream tasks. To better exploit the generic representation, we pre-train VL-BERT on the massive-scale Conceptual Captions dataset, together with text-only corpus. Extensive empirical analysis demonstrates that the pre-training procedure can better align the visual-linguistic clues and benefit the downstream tasks, such as visual commonsense reasoning, visual question answering and referring expression comprehension. It is worth noting that VL-BERT achieved the first place of single model on the leaderboard of the VCR benchmark. Code is released at \\url{https://github.com/jackroos/VL-BERT}.","url_abs":"https://arxiv.org/abs/1908.08530v4","url_pdf":"https://arxiv.org/pdf/1908.08530v4.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":"vl-bert-pre-training-of-generic-visual","repo_url":"https://github.com/jackroos/VL-BERT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"vl-bert-pre-training-of-generic-visual","repo_url":"https://github.com/ImperialNLP/BertGen","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"vl-bert-pre-training-of-generic-visual","repo_url":"https://github.com/jules-samaran/vl-bert","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-text-matching","task_name":"Image-text matching"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"referring-expression","task_name":"Referring Expression"},{"task_slug":"referring-expression-comprehension","task_name":"Referring Expression Comprehension"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"visual-commonsense-reasoning","task_name":"Visual Commonsense Reasoning"},{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"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":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"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":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"},{"method_slug":"vl-bert","method_name":"VL-BERT"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[{"slug":"vl-bert","name":"VL-BERT","full_name":"Visual-Linguistic BERT"}],"results":[{"leaderboard":"/sota/image-text-matching-on-commercialadsdataset","task":"Image-text matching","dataset":"CommercialAdsDataset","model":"VL-BERT","rank_in_archive_order":5,"of":8,"metrics":{"ADD(S) AUC":"86.27"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-a-dev","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-A) dev","model":"VL-BERTLARGE","rank_in_archive_order":1,"of":3,"metrics":{"Accuracy":"75.5"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-a-dev","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-A) dev","model":"VL-BERTBASE","rank_in_archive_order":2,"of":3,"metrics":{"Accuracy":"73.8"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-a-test","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-A) test","model":"VL-BERTLARGE","rank_in_archive_order":7,"of":11,"metrics":{"Accuracy":"75.8"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-ar-dev","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-AR) dev","model":"VL-BERTLARGE","rank_in_archive_order":1,"of":3,"metrics":{"Accuracy":"58.9"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-ar-dev","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-AR) dev","model":"VL-BERTBASE","rank_in_archive_order":2,"of":3,"metrics":{"Accuracy":"55.2"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-ar-test","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-AR) test","model":"VL-BERTLARGE","rank_in_archive_order":5,"of":7,"metrics":{"Accuracy":"59.7"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-qa-r-dev","task":"Visual Question Answering (VQA)","dataset":"VCR (QA-R) dev","model":"VL-BERTLARGE","rank_in_archive_order":1,"of":3,"metrics":{"Accuracy":"77.9"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-qa-r-dev","task":"Visual Question Answering (VQA)","dataset":"VCR (QA-R) dev","model":"VL-BERTBASE","rank_in_archive_order":2,"of":3,"metrics":{"Accuracy":"74.4"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-qa-r-test","task":"Visual Question Answering (VQA)","dataset":"VCR (QA-R) test","model":"VL-BERTLARGE","rank_in_archive_order":6,"of":8,"metrics":{"Accuracy":"78.4"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vqa-v2-test-dev","task":"Visual Question Answering (VQA)","dataset":"VQA v2 test-dev","model":"VL-BERTLARGE","rank_in_archive_order":24,"of":56,"metrics":{"Accuracy":"71.79"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vqa-v2-test-dev","task":"Visual Question Answering (VQA)","dataset":"VQA v2 test-dev","model":"VL-BERTBASE","rank_in_archive_order":27,"of":56,"metrics":{"Accuracy":"71.16"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vqa-v2-test-std","task":"Visual Question Answering (VQA)","dataset":"VQA v2 test-std","model":"VL-BERTLARGE","rank_in_archive_order":22,"of":38,"metrics":{"overall":"72.2"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1908.08530","atlas_url":"https://app.syntology.ai/?focus=1908.08530","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1908.08530"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/jackroos/VL-BERT","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ImperialNLP/BertGen","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jules-samaran/vl-bert","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"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":"a905f7851b56e39c","entry":"to_cuda","repo":"jackroos/VL-BERT","repo_kind":"official","path":"common/trainer.py","file_url":"https://github.com/jackroos/VL-BERT/blob/HEAD/common/trainer.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":"a905f7851b56e39c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}