Metraipolis › archive › dc_72bfaebe7afc

What Does It Mean to Forget a Person? Individual-Level Unlearning in Vision-Language Models — arXiv:2609.33481

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33481
License (as archived)
CC0-1.0
Archived
2026-10-07T22:13:45.357Z
SHA-256 of the archived text
dbeb916f129386c60f4972433474604a7fccc2115c54704748b77ba5cf2402cb
Stable URL of this page
https://metraipolis.com/documento/dc_72bfaebe7afc

How to cite

What Does It Mean to Forget a Person? Individual-Level Unlearning in Vision-Language Models — arXiv:2609.33481 — arXiv (en), https://arxiv.org/abs/2609.33481 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_72bfaebe7afc (Metraipolis archive dc_72bfaebe7afc, sha256 dbeb916f129386c60f4972433474604a7fccc2115c54704748b77ba5cf2402cb)
BibTeX
@misc{dc72bfaebe7afc,
  title = {What Does It Mean to Forget a Person? Individual-Level Unlearning in Vision-Language Models — arXiv:2609.33481},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33481},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_72bfaebe7afc; sha256 dbeb916f129386c60f4972433474604a7fccc2115c54704748b77ba5cf2402cb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33481. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Erasing individual identities from Vision-Language Models (VLMs) is uniquely challenging because personal data is entangled across modalities rather than stored as isolated attributes. However, existing multimodal unlearning benchmarks primarily evaluate attribute-centric forgetting, overlooking the more critical objective of individual-level unlearning: eliminating a model's ability to access, link, and reconstruct target-related information across modalities. To address this gap, we propose IDUnlearn-Bench, the first benchmark for individual-level multimodal unlearning in VLMs. It represents each individual as connected multimodal evidence and evaluates four task families: attribute access, identity access, identity binding, and identity reconstruction. Experiments on representative VLMs and unlearning methods show that successful attribute-centric forgetting often leaves substantial identity-level knowledge intact and can be non-monotonic: reducing one form of risk may amplify another. Models may suppress selected information while still identifying the target, linking records, or reconstructing the individual. These findings reveal a fundamental gap between forgetting information about a person and forgetting the person as a whole.

Tema degli esploratori: governance digitale e beni comuni (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CY).
Categorie arXiv: Cryptography and Security (cs.CR) ; Computers and Society (cs.CY).
Pagina del lavoro: https://arxiv.org/abs/2609.33481 — 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 2026-10-07T21:55:04.265Z dalla pagina https://arxiv.org/catchup/cs.CY/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).