{"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/imagebert-cross-modal-pre-training-with-large","title":"ImageBERT: Cross-modal Pre-training with Large-scale Weak-supervised Image-Text Data","arxiv_id":"2001.07966","date":"2020-01-22","proceeding":null,"authors":["Di Qi","Lin Su","Jia Song","Edward Cui","Taroon Bharti","Arun Sacheti"],"abstract":"In this paper, we introduce a new vision-language pre-trained model -- ImageBERT -- for image-text joint embedding. Our model is a Transformer-based model, which takes different modalities as input and models the relationship between them. The model is pre-trained on four tasks simultaneously: Masked Language Modeling (MLM), Masked Object Classification (MOC), Masked Region Feature Regression (MRFR), and Image Text Matching (ITM). To further enhance the pre-training quality, we have collected a Large-scale weAk-supervised Image-Text (LAIT) dataset from Web. We first pre-train the model on this dataset, then conduct a second stage pre-training on Conceptual Captions and SBU Captions. Our experiments show that multi-stage pre-training strategy outperforms single-stage pre-training. We also fine-tune and evaluate our pre-trained ImageBERT model on image retrieval and text retrieval tasks, and achieve new state-of-the-art results on both MSCOCO and Flickr30k datasets.","url_abs":"https://arxiv.org/abs/2001.07966v2","url_pdf":"https://arxiv.org/pdf/2001.07966v2.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":[],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"image-text-matching","task_name":"Image-text matching"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"masked-language-modeling","task_name":"Masked Language Modeling"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"text-matching","task_name":"Text Matching"},{"task_slug":"text-retrieval","task_name":"Text Retrieval"},{"task_slug":"zero-shot-cross-modal-retrieval","task_name":"Zero-Shot Cross-Modal Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/zero-shot-cross-modal-retrieval-on-coco-2014","task":"Zero-Shot Cross-Modal Retrieval","dataset":"COCO 2014","model":"ImageBERT","rank_in_archive_order":17,"of":18,"metrics":{"Image-to-text R@1":"44.0","Image-to-text R@10":"80.4","Image-to-text R@5":"71.2","Text-to-image R@1":"32.3","Text-to-image R@10":"70.2","Text-to-image R@5":"59.0"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-cross-modal-retrieval-on-flickr30k","task":"Zero-Shot Cross-Modal Retrieval","dataset":"Flickr30k","model":"ImageBERT","rank_in_archive_order":20,"of":22,"metrics":{"Image-to-text R@1":"70.7","Image-to-text R@10":"94.0","Image-to-text R@5":"90.2","Text-to-image R@1":"54.3","Text-to-image R@10":"87.5","Text-to-image R@5":"79.6"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2001.07966","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}