{"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/cafe-catastrophic-data-leakage-in-vertical","title":"CAFE: Catastrophic Data Leakage in Vertical Federated Learning","arxiv_id":"2110.15122","date":"2021-10-26","proceeding":null,"authors":["Xiao Jin","Pin-Yu Chen","Chia-Yi Hsu","Chia-Mu Yu","Tianyi Chen"],"abstract":"Recent studies show that private training data can be leaked through the gradients sharing mechanism deployed in distributed machine learning systems, such as federated learning (FL). Increasing batch size to complicate data recovery is often viewed as a promising defense strategy against data leakage. In this paper, we revisit this defense premise and propose an advanced data leakage attack with theoretical justification to efficiently recover batch data from the shared aggregated gradients. We name our proposed method as catastrophic data leakage in vertical federated learning (CAFE). Comparing to existing data leakage attacks, our extensive experimental results on vertical FL settings demonstrate the effectiveness of CAFE to perform large-batch data leakage attack with improved data recovery quality. We also propose a practical countermeasure to mitigate CAFE. Our results suggest that private data participated in standard FL, especially the vertical case, have a high risk of being leaked from the training gradients. Our analysis implies unprecedented and practical data leakage risks in those learning settings. The code of our work is available at https://github.com/DeRafael/CAFE.","url_abs":"https://arxiv.org/abs/2110.15122v4","url_pdf":"https://arxiv.org/pdf/2110.15122v4.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":"cafe-catastrophic-data-leakage-in-vertical","repo_url":"https://github.com/derafael/cafe","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"federated-learning","task_name":"Federated Learning"},{"task_slug":"vertical-federated-learning","task_name":"Vertical Federated Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2110.15122","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2110.15122"}},"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/derafael/cafe","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":8},"by_repo_kind":{"official":{"samples":8,"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":"130ca908ed948ac1","entry":"cafe_middle_input","repo":"derafael/cafe","repo_kind":"official","path":"double_shot.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/double_shot.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":"130ca908ed948ac1"}},{"code_sha256_prefix":"d1b33a7ec5011fb2","entry":"cafe_middle_output_gradient","repo":"derafael/cafe","repo_kind":"official","path":"first_shot.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/first_shot.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":"d1b33a7ec5011fb2"}},{"code_sha256_prefix":"778b9ea3a727c31d","entry":"compute_accuracy","repo":"derafael/cafe","repo_kind":"official","path":"utils.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/utils.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":"778b9ea3a727c31d"}},{"code_sha256_prefix":"a7f22ab2c8e8dccd","entry":"compute_loss","repo":"derafael/cafe","repo_kind":"official","path":"utils.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/utils.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":"a7f22ab2c8e8dccd"}},{"code_sha256_prefix":"fc8c5619a11080bb","entry":"dummy_data_init","repo":"derafael/cafe","repo_kind":"official","path":"utils.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/utils.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":"fc8c5619a11080bb"}},{"code_sha256_prefix":"1b6b9fd7caf0e0f8","entry":"load_and_preprocess_image","repo":"derafael/cafe","repo_kind":"official","path":"data_preprocess.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/data_preprocess.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":"1b6b9fd7caf0e0f8"}},{"code_sha256_prefix":"0ca8b324119af42c","entry":"load_data","repo":"derafael/cafe","repo_kind":"official","path":"data_preprocess.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/data_preprocess.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":"0ca8b324119af42c"}},{"code_sha256_prefix":"7d871e6b3dc081d9","entry":"preprocess_image","repo":"derafael/cafe","repo_kind":"official","path":"data_preprocess.py","file_url":"https://github.com/derafael/cafe/blob/HEAD/data_preprocess.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":"7d871e6b3dc081d9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}