Metraipolis › archive › dc_70eb27cb68bd
The Nearest Target Is the Wrong One: Target Separation in Arc2Face Identity Unlearning — arXiv:2608.30087
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.30087
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T09:59:19.541Z
- SHA-256 of the archived text
5dc98525c69dd94c9d538e244251ec99477e59726ad9f897c419ad22bad5ec64- Stable URL of this page
https://metraipolis.com/documento/dc_70eb27cb68bd
How to cite
The Nearest Target Is the Wrong One: Target Separation in Arc2Face Identity Unlearning — arXiv:2608.30087 — arXiv (en), https://arxiv.org/abs/2608.30087 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_70eb27cb68bd (Metraipolis archive dc_70eb27cb68bd, sha256 5dc98525c69dd94c9d538e244251ec99477e59726ad9f897c419ad22bad5ec64)
BibTeX
@misc{dc70eb27cb68bd,
title = {The Nearest Target Is the Wrong One: Target Separation in Arc2Face Identity Unlearning — arXiv:2608.30087},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.30087},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_70eb27cb68bd; sha256 5dc98525c69dd94c9d538e244251ec99477e59726ad9f897c419ad22bad5ec64},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30087. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Unlearning an identity from a face-conditioned generator by redirecting its conditioning embedding can silently fail if the redirected output is still verified as the original person. We show that this failure depends on a controllable choice of how far the redirection target lies from the forget identity in recognition space, and that the most intuitive target, the nearest neighbour, is the one most likely to cause it. We audit Arc2Face with a locked ArcFace protocol and a projection adapter that redirects identity conditioning before generation. On a hard-neighbour stress test built from the hardest 0.5% of eligible identities, four target-selection policies show a monotonic response: clean forgetting rises from 9/30 groups under the nearest hard target to 30/30 under the least similar one. Mean forget-identity re-identification falls from 51.9 to 0.0 while mean retention stays flat. This reflects successful redirection rather than outputs becoming unverifiable: 710 of 720 least-sim-hard generations arrive at the chosen target, with no leakage to unrelated identities. Re-verifying identical images with an independent recogniser (AdaFace) preserves that trend, correlating at r=0.94, arguing against a verifier artefact. Target separation is thus a first-order, reportable design variable for identity unlearning. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-01 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.30087 — 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 2026-10-17T09:28:48.108Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).