{"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/erasing-without-remembering-safeguarding","title":"Erasing Without Remembering: Implicit Knowledge Forgetting in Large Language Models","arxiv_id":"2502.19982","date":"2025-02-27","proceeding":null,"authors":["Huazheng Wang","Yongcheng Jing","Haifeng Sun","Yingjie Wang","Jingyu Wang","Jianxin Liao","DaCheng Tao"],"abstract":"In this paper, we investigate knowledge forgetting in large language models with a focus on its generalisation--ensuring that models forget not only specific training samples but also related implicit knowledge. To this end, we begin by identifying a broader unlearning scope that includes both target data and logically associated samples, including rephrased, subject-replaced, one-hop reasoned, and relation-reversed data. To rigorously evaluate generalisation, we introduce UGBench, the first comprehensive benchmark specifically designed to assess the unlearning of in-scope implicit knowledge covering 13 state-of-the-art methods across three datasets. UGBench reveals that unlearned models can still recall paraphrased answers and retain target facts in intermediate layers. This motivates us to take a preliminary step toward more generalised implicit knowledge forgetting by proposing PerMU, a novel probability perturbation-based unlearning paradigm. PerMU simulates adversarial unlearning samples to eliminate fact-related tokens from the logit distribution, collectively reducing the probabilities of all answer-associated tokens. Experiments are conducted on a diverse range of datasets, including TOFU, Harry Potter, ZsRE, WMDP, and MUSE, using models ranging from 1.3B to 13B in scale. The results demonstrate that PerMU delivers up to a 50.40% improvement in unlearning vanilla target data while maintaining a 40.73% boost in forgetting implicit knowledge. Our code can be found in https://github.com/MaybeLizzy/UGBench.","url_abs":"https://arxiv.org/abs/2502.19982v2","url_pdf":"https://arxiv.org/pdf/2502.19982v2.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":"erasing-without-remembering-safeguarding","repo_url":"https://github.com/maybelizzy/ugbench","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-unlearning","task_name":"Machine Unlearning"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2502.19982","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}