{"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/papillon-privacy-preservation-from-internet","title":"PAPILLON: Privacy Preservation from Internet-based and Local Language Model Ensembles","arxiv_id":"2410.17127","date":"2024-10-22","proceeding":null,"authors":["Li Siyan","Vethavikashini Chithrra Raghuram","Omar Khattab","Julia Hirschberg","Zhou Yu"],"abstract":"Users can divulge sensitive information to proprietary LLM providers, raising significant privacy concerns. While open-source models, hosted locally on the user's machine, alleviate some concerns, models that users can host locally are often less capable than proprietary frontier models. Toward preserving user privacy while retaining the best quality, we propose Privacy-Conscious Delegation, a novel task for chaining API-based and local models. We utilize recent public collections of user-LLM interactions to construct a natural benchmark called PUPA, which contains personally identifiable information (PII). To study potential approaches, we devise PAPILLON, a multi-stage LLM pipeline that uses prompt optimization to address a simpler version of our task. Our best pipeline maintains high response quality for 85.5% of user queries while restricting privacy leakage to only 7.5%. We still leave a large margin to the generation quality of proprietary LLMs for future work. Our data and code is available at https://github.com/siyan-sylvia-li/PAPILLON.","url_abs":"https://arxiv.org/abs/2410.17127v3","url_pdf":"https://arxiv.org/pdf/2410.17127v3.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":"papillon-privacy-preservation-from-internet","repo_url":"https://github.com/columbia-nlp-lab/papillon","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"papillon-privacy-preservation-from-internet","repo_url":"https://github.com/siyan-sylvia-li/papillon","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2410.17127","atlas_url":"https://app.syntology.ai/?focus=2410.17127","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.17127"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/columbia-nlp-lab/papillon","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/siyan-sylvia-li/papillon","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":5},"by_repo_kind":{"official":{"samples":5,"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":"7bbb3a6a63326205","entry":"context_independence","repo":"columbia-nlp-lab/papillon","repo_kind":"official","path":"pupa/filter_context_dependence.py","file_url":"https://github.com/columbia-nlp-lab/papillon/blob/HEAD/pupa/filter_context_dependence.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":"7bbb3a6a63326205"}},{"code_sha256_prefix":"cb77a6c5bc751791","entry":"generate_extract","repo":"columbia-nlp-lab/papillon","repo_kind":"official","path":"pupa/create_privacy_span.py","file_url":"https://github.com/columbia-nlp-lab/papillon/blob/HEAD/pupa/create_privacy_span.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":"cb77a6c5bc751791"}},{"code_sha256_prefix":"edd8a46261673148","entry":"parse_model_prompt","repo":"columbia-nlp-lab/papillon","repo_kind":"official","path":"papillon/evaluate_papillon.py","file_url":"https://github.com/columbia-nlp-lab/papillon/blob/HEAD/papillon/evaluate_papillon.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":"edd8a46261673148"}},{"code_sha256_prefix":"4d78000412bfa8cb","entry":"process_scores","repo":"columbia-nlp-lab/papillon","repo_kind":"official","path":"papillon/llm_judge.py","file_url":"https://github.com/columbia-nlp-lab/papillon/blob/HEAD/papillon/llm_judge.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":"4d78000412bfa8cb"}},{"code_sha256_prefix":"e37bc98bd166fd1f","entry":"unredact_information","repo":"columbia-nlp-lab/papillon","repo_kind":"official","path":"pupa/create_privacy_span.py","file_url":"https://github.com/columbia-nlp-lab/papillon/blob/HEAD/pupa/create_privacy_span.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":"e37bc98bd166fd1f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}