{"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/offsite-tuning-transfer-learning-without-full","title":"Offsite-Tuning: Transfer Learning without Full Model","arxiv_id":"2302.04870","date":"2023-02-09","proceeding":null,"authors":["Guangxuan Xiao","Ji Lin","Song Han"],"abstract":"Transfer learning is important for foundation models to adapt to downstream tasks. However, many foundation models are proprietary, so users must share their data with model owners to fine-tune the models, which is costly and raise privacy concerns. Moreover, fine-tuning large foundation models is computation-intensive and impractical for most downstream users. In this paper, we propose Offsite-Tuning, a privacy-preserving and efficient transfer learning framework that can adapt billion-parameter foundation models to downstream data without access to the full model. In offsite-tuning, the model owner sends a light-weight adapter and a lossy compressed emulator to the data owner, who then fine-tunes the adapter on the downstream data with the emulator's assistance. The fine-tuned adapter is then returned to the model owner, who plugs it into the full model to create an adapted foundation model. Offsite-tuning preserves both parties' privacy and is computationally more efficient than the existing fine-tuning methods that require access to the full model weights. We demonstrate the effectiveness of offsite-tuning on various large language and vision foundation models. Offsite-tuning can achieve comparable accuracy as full model fine-tuning while being privacy-preserving and efficient, achieving 6.5x speedup and 5.6x memory reduction. Code is available at https://github.com/mit-han-lab/offsite-tuning.","url_abs":"https://arxiv.org/abs/2302.04870v1","url_pdf":"https://arxiv.org/pdf/2302.04870v1.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":"offsite-tuning-transfer-learning-without-full","repo_url":"https://github.com/mit-han-lab/offsite-tuning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"privacy-preserving","task_name":"Privacy Preserving"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"model","task_name":"model"}],"methods":[{"method_slug":"adapter","method_name":"Adapter"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2302.04870","atlas_url":"https://app.syntology.ai/?focus=2302.04870","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2302.04870"}},"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/mit-han-lab/offsite-tuning","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":4,"unverified":3},"by_repo_kind":{"official":{"samples":7,"ran":4,"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":"644daf7343d41357","entry":"add_epilogue","repo":"mit-han-lab/offsite-tuning","repo_kind":"official","path":"offsite_tuning/utils.py","file_url":"https://github.com/mit-han-lab/offsite-tuning/blob/HEAD/offsite_tuning/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"644daf7343d41357"}},{"code_sha256_prefix":"4c9ea68663be734c","entry":"add_prologue","repo":"mit-han-lab/offsite-tuning","repo_kind":"official","path":"offsite_tuning/utils.py","file_url":"https://github.com/mit-han-lab/offsite-tuning/blob/HEAD/offsite_tuning/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4c9ea68663be734c"}},{"code_sha256_prefix":"e503e07fecd0ae48","entry":"get_lm_datasets","repo":"mit-han-lab/offsite-tuning","repo_kind":"official","path":"offsite_tuning/data.py","file_url":"https://github.com/mit-han-lab/offsite-tuning/blob/HEAD/offsite_tuning/data.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e503e07fecd0ae48"}},{"code_sha256_prefix":"0ed6b0ea7d0d953a","entry":"map_dataset_name_and_config","repo":"mit-han-lab/offsite-tuning","repo_kind":"official","path":"offsite_tuning/tasks.py","file_url":"https://github.com/mit-han-lab/offsite-tuning/blob/HEAD/offsite_tuning/tasks.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0ed6b0ea7d0d953a"}},{"code_sha256_prefix":"f2c5c52ee0e49837","entry":"add_adapter","repo":"mit-han-lab/offsite-tuning","repo_kind":"official","path":"offsite_tuning/param_efficient.py","file_url":"https://github.com/mit-han-lab/offsite-tuning/blob/HEAD/offsite_tuning/param_efficient.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":"f2c5c52ee0e49837"}},{"code_sha256_prefix":"ebd8572c5c5152b0","entry":"get_tokenized_datasets","repo":"mit-han-lab/offsite-tuning","repo_kind":"official","path":"offsite_tuning/data.py","file_url":"https://github.com/mit-han-lab/offsite-tuning/blob/HEAD/offsite_tuning/data.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":"ebd8572c5c5152b0"}},{"code_sha256_prefix":"0a5252c9d50f4927","entry":"uniform_choose_layers","repo":"mit-han-lab/offsite-tuning","repo_kind":"official","path":"offsite_tuning/utils.py","file_url":"https://github.com/mit-han-lab/offsite-tuning/blob/HEAD/offsite_tuning/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":"0a5252c9d50f4927"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}