{"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/metapoison-practical-general-purpose-clean","title":"MetaPoison: Practical General-purpose Clean-label Data Poisoning","arxiv_id":"2004.00225","date":"2020-04-01","proceeding":"NeurIPS 2020 12","authors":["W. Ronny Huang","Jonas Geiping","Liam Fowl","Gavin Taylor","Tom Goldstein"],"abstract":"Data poisoning -- the process by which an attacker takes control of a model by making imperceptible changes to a subset of the training data -- is an emerging threat in the context of neural networks. Existing attacks for data poisoning neural networks have relied on hand-crafted heuristics, because solving the poisoning problem directly via bilevel optimization is generally thought of as intractable for deep models. We propose MetaPoison, a first-order method that approximates the bilevel problem via meta-learning and crafts poisons that fool neural networks. MetaPoison is effective: it outperforms previous clean-label poisoning methods by a large margin. MetaPoison is robust: poisoned data made for one model transfer to a variety of victim models with unknown training settings and architectures. MetaPoison is general-purpose, it works not only in fine-tuning scenarios, but also for end-to-end training from scratch, which till now hasn't been feasible for clean-label attacks with deep nets. MetaPoison can achieve arbitrary adversary goals -- like using poisons of one class to make a target image don the label of another arbitrarily chosen class. Finally, MetaPoison works in the real-world. We demonstrate for the first time successful data poisoning of models trained on the black-box Google Cloud AutoML API. Code and premade poisons are provided at https://github.com/wronnyhuang/metapoison","url_abs":"https://arxiv.org/abs/2004.00225v2","url_pdf":"https://arxiv.org/pdf/2004.00225v2.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":"metapoison-practical-general-purpose-clean","repo_url":"https://github.com/wronnyhuang/metapoison","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"metapoison-practical-general-purpose-clean","repo_url":"https://github.com/JonasGeiping/poisoning-gradient-matching","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"automl","task_name":"AutoML"},{"task_slug":"bilevel-optimization","task_name":"Bilevel Optimization"},{"task_slug":"data-poisoning","task_name":"Data Poisoning"},{"task_slug":"meta-learning","task_name":"Meta-Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2004.00225","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2004.00225"}},"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/wronnyhuang/metapoison","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/JonasGeiping/poisoning-gradient-matching","reach":null}],"summary":{"ran":2,"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1},"listed":{"samples":2,"ran":2,"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":3,"samples":[{"code_sha256_prefix":"e458f01331f5d8fb","entry":"MetaMonkey","repo":"JonasGeiping/poisoning-gradient-matching","repo_kind":"listed","path":"forest/witchcoven/witch_metapoison.py","file_url":"https://github.com/JonasGeiping/poisoning-gradient-matching/blob/HEAD/forest/witchcoven/witch_metapoison.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"LGPL-2.1","inline_ok":false,"mcp_get_code":{"code_sha256":"e458f01331f5d8fb"}},{"code_sha256_prefix":"a87097c3d8cc8031","entry":"WitchMetaPoison","repo":"JonasGeiping/poisoning-gradient-matching","repo_kind":"listed","path":"forest/witchcoven/witch_metapoison.py","file_url":"https://github.com/JonasGeiping/poisoning-gradient-matching/blob/HEAD/forest/witchcoven/witch_metapoison.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"LGPL-2.1","inline_ok":false,"mcp_get_code":{"code_sha256":"a87097c3d8cc8031"}},{"code_sha256_prefix":"f72746c04360b1ff","entry":"Meta","repo":"wronnyhuang/metapoison","repo_kind":"official","path":"meta.py","file_url":"https://github.com/wronnyhuang/metapoison/blob/HEAD/meta.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f72746c04360b1ff"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}