{"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/seps-a-separability-measure-for-robust","title":"SEPS: A Separability Measure for Robust Unlearning in LLMs","arxiv_id":"2505.14832","date":"2025-05-20","proceeding":null,"authors":["Wonje Jeung","Sangyeon Yoon","Albert No"],"abstract":"Machine unlearning aims to selectively remove targeted knowledge from Large Language Models (LLMs), ensuring they forget specified content while retaining essential information. Existing unlearning metrics assess whether a model correctly answers retain queries and rejects forget queries, but they fail to capture real-world scenarios where forget queries rarely appear in isolation. In fact, forget and retain queries often coexist within the same prompt, making mixed-query evaluation crucial. We introduce SEPS, an evaluation framework that explicitly measures a model's ability to both forget and retain information within a single prompt. Through extensive experiments across three benchmarks, we identify two key failure modes in existing unlearning methods: (1) untargeted unlearning indiscriminately erases both forget and retain content once a forget query appears, and (2) targeted unlearning overfits to single-query scenarios, leading to catastrophic failures when handling multiple queries. To address these issues, we propose Mixed Prompt (MP) unlearning, a strategy that integrates both forget and retain queries into a unified training objective. Our approach significantly improves unlearning effectiveness, demonstrating robustness even in complex settings with up to eight mixed forget and retain queries in a single prompt.","url_abs":"https://arxiv.org/abs/2505.14832v1","url_pdf":"https://arxiv.org/pdf/2505.14832v1.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":[],"tasks":[{"task_slug":"machine-unlearning","task_name":"Machine Unlearning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2505.14832","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2505.14832"}},"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":"deterministic:regex_extraction","url":"https://github.com/AI-ISL/SEPS","reach":null}],"summary":{"unverified":3},"by_repo_kind":{"found_in_text":{"samples":3,"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":3,"samples":[{"code_sha256_prefix":"9ab3eb63651dacc3","entry":"find_all_token_indices_batch1","repo":"AI-ISL/SEPS","repo_kind":"found_in_text","path":"trainer/losses.py","file_url":"https://github.com/AI-ISL/SEPS/blob/HEAD/trainer/losses.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":"9ab3eb63651dacc3"}},{"code_sha256_prefix":"27b799822f45e087","entry":"get_mixed_untargeted_loss1","repo":"AI-ISL/SEPS","repo_kind":"found_in_text","path":"trainer/losses.py","file_url":"https://github.com/AI-ISL/SEPS/blob/HEAD/trainer/losses.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":"27b799822f45e087"}},{"code_sha256_prefix":"f420314bd6faa95c","entry":"mixed_fr_untargeted_kl_loss","repo":"AI-ISL/SEPS","repo_kind":"found_in_text","path":"trainer/losses.py","file_url":"https://github.com/AI-ISL/SEPS/blob/HEAD/trainer/losses.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":"f420314bd6faa95c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}