{"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/soft-prompting-for-unlearning-in-large","title":"Soft Prompting for Unlearning in Large Language Models","arxiv_id":"2406.12038","date":"2024-06-17","proceeding":null,"authors":["Karuna Bhaila","Minh-Hao Van","Xintao Wu"],"abstract":"The widespread popularity of Large Language Models (LLMs), partly due to their unique ability to perform in-context learning, has also brought to light the importance of ethical and safety considerations when deploying these pre-trained models. In this work, we focus on investigating machine unlearning for LLMs motivated by data protection regulations. In contrast to the growing literature on fine-tuning methods to achieve unlearning, we focus on a comparatively lightweight alternative called soft prompting to realize the unlearning of a subset of training data. With losses designed to enforce forgetting as well as utility preservation, our framework \\textbf{S}oft \\textbf{P}rompting for \\textbf{U}n\\textbf{l}earning (SPUL) learns prompt tokens that can be appended to an arbitrary query to induce unlearning of specific examples at inference time without updating LLM parameters. We conduct a rigorous evaluation of the proposed method and our results indicate that SPUL can significantly improve the trade-off between utility and forgetting in the context of text classification and question answering with LLMs. We further validate our method using multiple LLMs to highlight the scalability of our framework and provide detailed insights into the choice of hyperparameters and the influence of the size of unlearning data. Our implementation is available at \\url{https://github.com/karuna-bhaila/llm_unlearning}.","url_abs":"https://arxiv.org/abs/2406.12038v2","url_pdf":"https://arxiv.org/pdf/2406.12038v2.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":"soft-prompting-for-unlearning-in-large","repo_url":"https://github.com/karuna-bhaila/llm_unlearning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"},{"task_slug":"machine-unlearning","task_name":"Machine Unlearning"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"text-classification","task_name":"Text Classification"},{"task_slug":"text-classification-1","task_name":"text-classification"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2406.12038","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.12038"}},"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/karuna-bhaila/llm_unlearning","reach":{"status":"ok"}}],"summary":{"ran":2,"ran_honours":1,"unverified":5},"by_repo_kind":{"official":{"samples":8,"ran":3,"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":8,"samples":[{"code_sha256_prefix":"49bb87ecc41d23fe","entry":"compute_metrics","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"inference_vanilla.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/inference_vanilla.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"49bb87ecc41d23fe"}},{"code_sha256_prefix":"39181472211f0ed1","entry":"get_model","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"inference_vanilla.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/inference_vanilla.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"39181472211f0ed1"}},{"code_sha256_prefix":"d2895bdf14e8f94f","entry":"preprocess_logits_for_metrics","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"utils.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d2895bdf14e8f94f"}},{"code_sha256_prefix":"7e1ff286a54b3ec5","entry":"get_data_path","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"utils.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/utils.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":"7e1ff286a54b3ec5"}},{"code_sha256_prefix":"cfc673fbc4f72125","entry":"get_logits_from_base_model","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"utils.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/utils.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":"cfc673fbc4f72125"}},{"code_sha256_prefix":"bb2d30c2f169cb27","entry":"get_lora_model","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"inference_base_model.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/inference_base_model.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":"bb2d30c2f169cb27"}},{"code_sha256_prefix":"7cf32abedf8023cd","entry":"get_lora_model","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"qlora.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/qlora.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":"7cf32abedf8023cd"}},{"code_sha256_prefix":"c9037aad77685aff","entry":"get_unlearn_dataset","repo":"karuna-bhaila/llm_unlearning","repo_kind":"official","path":"inference_vanilla.py","file_url":"https://github.com/karuna-bhaila/llm_unlearning/blob/HEAD/inference_vanilla.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":"c9037aad77685aff"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}