Metraipolis › archive › dc_7cb2cb18235c
Behavioral Forgetting Is Not Privacy in Machine Unlearning — arXiv:2610.21799
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.21799
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T10:01:51.417Z
- SHA-256 of the archived text
12885c02ad44c01911b4261e9915e4bd32ca0b3477a762e84da6fe9b72414752- Stable URL of this page
https://metraipolis.com/documento/dc_7cb2cb18235c
How to cite
Behavioral Forgetting Is Not Privacy in Machine Unlearning — arXiv:2610.21799 — arXiv (en), https://arxiv.org/abs/2610.21799 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7cb2cb18235c (Metraipolis archive dc_7cb2cb18235c, sha256 12885c02ad44c01911b4261e9915e4bd32ca0b3477a762e84da6fe9b72414752)
BibTeX
@misc{dc7cb2cb18235c,
title = {Behavioral Forgetting Is Not Privacy in Machine Unlearning — arXiv:2610.21799},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.21799},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7cb2cb18235c; sha256 12885c02ad44c01911b4261e9915e4bd32ca0b3477a762e84da6fe9b72414752},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21799. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Machine unlearning aims to remove targeted information from large language models, but existing evaluations often measure forgetting only on a designated benchmark distribution. We argue that such behavioral forgetting is an insufficient standard for privacy-oriented unlearning, and that evaluation should instead test whether forgetting persists across the access paths through which the same information may be recovered. We demonstrate this with a controlled case study of cross-lingual querying: across multiple multilingual model families, forgetting varies substantially when semantically equivalent information is queried in different languages and scripts. A single access path is therefore enough to expose a real gap between behavioral forgetting and information inaccessibility, motivating broader evaluation of access-path generalization and adversarial recoverability alongside in-distribution forgetting. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.21799 — 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 2026-10-16T08:44:47.675Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).