Metraipolis › archive › dc_fcab9e8a7222

Continual Data Unlearning in Diffusion Models via Transition-based Regularization — arXiv:2609.32328

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32328
License (as archived)
CC0-1.0
Archived
2026-10-07T22:36:30.075Z
SHA-256 of the archived text
ade66d139bbf88ccb93ecc2ce96f41345331c69e9a9d7110a7b14b29354eb31c
Stable URL of this page
https://metraipolis.com/documento/dc_fcab9e8a7222

How to cite

Continual Data Unlearning in Diffusion Models via Transition-based Regularization — arXiv:2609.32328 — arXiv (en), https://arxiv.org/abs/2609.32328 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fcab9e8a7222 (Metraipolis archive dc_fcab9e8a7222, sha256 ade66d139bbf88ccb93ecc2ce96f41345331c69e9a9d7110a7b14b29354eb31c)
BibTeX
@misc{dcfcab9e8a7222,
  title = {Continual Data Unlearning in Diffusion Models via Transition-based Regularization — arXiv:2609.32328},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32328},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fcab9e8a7222; sha256 ade66d139bbf88ccb93ecc2ce96f41345331c69e9a9d7110a7b14b29354eb31c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32328. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Data unlearning in diffusion models aims to remove the influence of specific training examples without suppressing the broader concepts they represent. However, when deletion requests arrive sequentially, updates for new requests can degrade generative utility and undermine earlier deletions. We propose a continual data unlearning framework that uses completed deletion transitions as directional references to regularize future updates. For each request, we record changes in denoiser responses on the same fixed noisy inputs before and after unlearning. Rather than matching full post-deletion responses, we apply a one-sided penalty that discourages reversal along the recorded directions relative to the post-deletion references, while leaving orthogonal response changes and progress beyond these references unpenalized. To keep storage independent of the number of requests, we maintain a fixed-capacity bank of representative transition records. Records are selected based on the local sensitivity of progress along their recorded directions to parameter updates, allowing them to be retained even when their penalties are inactive. Empirical evaluations show that the proposed framework achieves a better balance between deletion persistence and generative utility than existing unlearning baselines as requests accumulate, using only a small transition memory.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.32328 — 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 2026-10-07T22:14:27.658Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).