Metraipolis › archive › dc_4fbfae15d5b2
How Learning Governs Unlearning across the Memorization-Generalization Spectrum — arXiv:2610.08577
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.08577
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:21:59.798Z
- SHA-256 of the archived text
3f13b9f2c9d1fc9803fc2e0a2c8381a25e99f882f4082326a5c2df00a709bafc- Stable URL of this page
https://metraipolis.com/documento/dc_4fbfae15d5b2
How to cite
How Learning Governs Unlearning across the Memorization-Generalization Spectrum — arXiv:2610.08577 — arXiv (en), https://arxiv.org/abs/2610.08577 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4fbfae15d5b2 (Metraipolis archive dc_4fbfae15d5b2, sha256 3f13b9f2c9d1fc9803fc2e0a2c8381a25e99f882f4082326a5c2df00a709bafc)
BibTeX
@misc{dc4fbfae15d5b2,
title = {How Learning Governs Unlearning across the Memorization-Generalization Spectrum — arXiv:2610.08577},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.08577},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4fbfae15d5b2; sha256 3f13b9f2c9d1fc9803fc2e0a2c8381a25e99f882f4082326a5c2df00a709bafc},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08577. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
While unlearning seeks to negate undesired capabilities acquired through learning, little research has examined how the way models learn shapes their subsequent unlearning. In this paper, we investigate this connection from the perspectives of memorization and generalization, the two most representative yet competing strategies that models employ during training. We first classify memorization- and generalization-heavy models using grokking in modular addition and compare their responses to unlearning, showing that the latter suffer greater retain damage, i.e., a larger performance drop on the retain set. Furthermore, we conduct a finer-grained analysis by introducing bucketed modular addition, in which the respective contributions of the two strategies can be explicitly controlled across the memorization-generalization spectrum. In this setup, we reaffirm that the same trend persists and is nearly monotonic. We further demonstrate that this relationship also holds in LLM unlearning across verbatim and factual recall settings. Finally, we provide two practical insights for developing better unlearning methods, highlighting the importance of accounting for learning dynamics in unlearning. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.08577 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T11:15:24.848Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).