{"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/zero-and-r2d2-a-large-scale-chinese-cross","title":"CCMB: A Large-scale Chinese Cross-modal Benchmark","arxiv_id":"2205.03860","date":"2022-05-08","proceeding":null,"authors":["Chunyu Xie","Heng Cai","Jincheng Li","Fanjing Kong","Xiaoyu Wu","Jianfei Song","Henrique Morimitsu","Lin Yao","Dexin Wang","Xiangzheng Zhang","Dawei Leng","Baochang Zhang","Xiangyang Ji","Yafeng Deng"],"abstract":"Vision-language pre-training (VLP) on large-scale datasets has shown premier performance on various downstream tasks. In contrast to plenty of available benchmarks with English corpus, large-scale pre-training datasets and downstream datasets with Chinese corpus remain largely unexplored. In this work, we build a large-scale high-quality Chinese Cross-Modal Benchmark named CCMB for the research community, which contains the currently largest public pre-training dataset Zero and five human-annotated fine-tuning datasets for downstream tasks. Zero contains 250 million images paired with 750 million text descriptions, plus two of the five fine-tuning datasets are also currently the largest ones for Chinese cross-modal downstream tasks. Along with the CCMB, we also develop a VLP framework named R2D2, applying a pre-Ranking + Ranking strategy to learn powerful vision-language representations and a two-way distillation method (i.e., target-guided Distillation and feature-guided Distillation) to further enhance the learning capability. With the Zero and the R2D2 VLP framework, we achieve state-of-the-art performance on twelve downstream datasets from five broad categories of tasks including image-text retrieval, image-text matching, image caption, text-to-image generation, and zero-shot image classification. The datasets, models, and codes are available at https://github.com/yuxie11/R2D2","url_abs":"https://arxiv.org/abs/2205.03860v6","url_pdf":"https://arxiv.org/pdf/2205.03860v6.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":"zero-and-r2d2-a-large-scale-chinese-cross","repo_url":"https://github.com/yuxie11/R2D2","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":"image-generation","task_name":"Image Generation"},{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"image-text-retrieval","task_name":"Image-text Retrieval"},{"task_slug":"image-text-matching","task_name":"Image-text matching"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"text-matching","task_name":"Text Matching"},{"task_slug":"text-retrieval","task_name":"Text Retrieval"},{"task_slug":"text-to-image-generation-1","task_name":"Text to Image Generation"},{"task_slug":"text-to-image-generation","task_name":"Text-to-Image Generation"},{"task_slug":"zero-shot-image-classification","task_name":"Zero-Shot Image Classification"},{"task_slug":"zero-shot-image-retrieval","task_name":"Zero-shot Image Retrieval"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"r2d2","method_name":"R2D2"}],"datasets_introduced":[{"slug":"flickr30k-cna","name":"Flickr30k-CNA","full_name":"Flickr30k-Chinese All"},{"slug":"icm","name":"ICM","full_name":"Image-Caption Matching Dataset"},{"slug":"icr","name":"ICR","full_name":"Image-Caption Retrieval Dataset"},{"slug":"iqm","name":"IQM","full_name":"Image-Query Matching Dataset"},{"slug":"image-query-retrieval-dataset","name":"IQR","full_name":"Image-Query Retrieval Dataset"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-retrieval-on-coco-cn","task":"Image Retrieval","dataset":"COCO-CN","model":"R2D2 (ViT-L/14)","rank_in_archive_order":3,"of":9,"metrics":{"R@1":"79.1","R@10":"98.9","R@5":"96.5"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-coco-cn","task":"Image Retrieval","dataset":"COCO-CN","model":"R2D2 (ViT-B)","rank_in_archive_order":6,"of":9,"metrics":{"R@1":"75.1","R@10":"98.1","R@5":"94.2"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-flickr30k-cn","task":"Image Retrieval","dataset":"Flickr30k-CN","model":"R2D2 (ViT-L/14)","rank_in_archive_order":4,"of":11,"metrics":{"R@1":"84.4","R@10":"98.4","R@5":"96.7"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-flickr30k-cn","task":"Image Retrieval","dataset":"Flickr30k-CN","model":"R2D2 (ViT-B)","rank_in_archive_order":8,"of":11,"metrics":{"R@1":"78.3","R@10":"97.0","R@5":"94.6"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-muge-retrieval","task":"Image Retrieval","dataset":"MUGE Retrieval","model":"R2D2 (ViT-L/14)","rank_in_archive_order":4,"of":9,"metrics":{"Mean Recall":"77.5","R@1":"60.1","R@10":"89.4","R@5":"82.9"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-muge-retrieval","task":"Image Retrieval","dataset":"MUGE Retrieval","model":"R2D2 (ViT-B)","rank_in_archive_order":8,"of":9,"metrics":{"Mean Recall":"68.7","R@1":"47.4","R@10":"83.5","R@5":"75.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2205.03860","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2205.03860"}},"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/yuxie11/R2D2","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":9},"by_repo_kind":{"official":{"samples":9,"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":"7bf8bfbf255a9e9a","entry":"compute_image_features","repo":"yuxie11/R2D2","repo_kind":"official","path":"train_r2d2_retrieval.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/train_r2d2_retrieval.py","link_basis":"harvester_set","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":"7bf8bfbf255a9e9a"}},{"code_sha256_prefix":"4914409d106d9149","entry":"compute_text_features","repo":"yuxie11/R2D2","repo_kind":"official","path":"train_r2d2_retrieval.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/train_r2d2_retrieval.py","link_basis":"harvester_set","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":"4914409d106d9149"}},{"code_sha256_prefix":"be3c5540222af5be","entry":"evaluation","repo":"yuxie11/R2D2","repo_kind":"official","path":"r2d2_inference_demo.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/r2d2_inference_demo.py","link_basis":"harvester_set","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":"be3c5540222af5be"}},{"code_sha256_prefix":"ff0f588c0f5b23d9","entry":"gather_with_grad","repo":"yuxie11/R2D2","repo_kind":"official","path":"models/r2d2_retrieval.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/models/r2d2_retrieval.py","link_basis":"harvester_set","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":"ff0f588c0f5b23d9"}},{"code_sha256_prefix":"d6695d6f18eb07ee","entry":"gather_with_no_grad","repo":"yuxie11/R2D2","repo_kind":"official","path":"models/r2d2_retrieval.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/models/r2d2_retrieval.py","link_basis":"harvester_set","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":"d6695d6f18eb07ee"}},{"code_sha256_prefix":"619d3e0687ecaf67","entry":"load_checkpoint","repo":"yuxie11/R2D2","repo_kind":"official","path":"models/r2d2.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/models/r2d2.py","link_basis":"harvester_set","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":"619d3e0687ecaf67"}},{"code_sha256_prefix":"494b925526e287c1","entry":"make_optimizer","repo":"yuxie11/R2D2","repo_kind":"official","path":"train_r2d2_matching.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/train_r2d2_matching.py","link_basis":"harvester_set","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":"494b925526e287c1"}},{"code_sha256_prefix":"6f7d7b0cd2149e78","entry":"prd2","repo":"yuxie11/R2D2","repo_kind":"official","path":"models/prd2.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/models/prd2.py","link_basis":"harvester_set","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":"6f7d7b0cd2149e78"}},{"code_sha256_prefix":"2ffab879ffc06bbe","entry":"preprocess","repo":"yuxie11/R2D2","repo_kind":"official","path":"r2d2_inference_demo.py","file_url":"https://github.com/yuxie11/R2D2/blob/HEAD/r2d2_inference_demo.py","link_basis":"harvester_set","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":"2ffab879ffc06bbe"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}