{"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/tmi-finetuned-models-leak-private-information","title":"TMI! Finetuned Models Leak Private Information from their Pretraining Data","arxiv_id":"2306.01181","date":"2023-06-01","proceeding":null,"authors":["John Abascal","Stanley Wu","Alina Oprea","Jonathan Ullman"],"abstract":"Transfer learning has become an increasingly popular technique in machine learning as a way to leverage a pretrained model trained for one task to assist with building a finetuned model for a related task. This paradigm has been especially popular for $\\textit{privacy}$ in machine learning, where the pretrained model is considered public, and only the data for finetuning is considered sensitive. However, there are reasons to believe that the data used for pretraining is still sensitive, making it essential to understand how much information the finetuned model leaks about the pretraining data. In this work we propose a new membership-inference threat model where the adversary only has access to the finetuned model and would like to infer the membership of the pretraining data. To realize this threat model, we implement a novel metaclassifier-based attack, $\\textbf{TMI}$, that leverages the influence of memorized pretraining samples on predictions in the downstream task. We evaluate $\\textbf{TMI}$ on both vision and natural language tasks across multiple transfer learning settings, including finetuning with differential privacy. Through our evaluation, we find that $\\textbf{TMI}$ can successfully infer membership of pretraining examples using query access to the finetuned model. An open-source implementation of $\\textbf{TMI}$ can be found on GitHub: https://github.com/johnmath/tmi-pets24.","url_abs":"https://arxiv.org/abs/2306.01181v3","url_pdf":"https://arxiv.org/pdf/2306.01181v3.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":"tmi-finetuned-models-leak-private-information","repo_url":"https://github.com/johnmath/tmi-pets24","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2306.01181","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2306.01181"}},"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/johnmath/tmi-pets24","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":2},"by_repo_kind":{"official":{"samples":2,"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":"7d23b04b8fbfa0d2","entry":"plot_rocs","repo":"johnmath/tmi-pets24","repo_kind":"official","path":"utils/transfermi.py","file_url":"https://github.com/johnmath/tmi-pets24/blob/HEAD/utils/transfermi.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":"7d23b04b8fbfa0d2"}},{"code_sha256_prefix":"ceb4a41d93e5cebc","entry":"sweep","repo":"johnmath/tmi-pets24","repo_kind":"official","path":"utils/transfermi.py","file_url":"https://github.com/johnmath/tmi-pets24/blob/HEAD/utils/transfermi.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":"ceb4a41d93e5cebc"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}