Metraipolis › archive › dc_0eea335d545a
SIEVE: Selective attention-value Suppression for Vision-Language Models Unlearning — arXiv:2610.01962
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.01962
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T19:16:58.208Z
- SHA-256 of the archived text
8f0595acd829de6439df837ae159ba82623df3c779c19cf00ae0424310b10d40- Stable URL of this page
https://metraipolis.com/documento/dc_0eea335d545a
How to cite
SIEVE: Selective attention-value Suppression for Vision-Language Models Unlearning — arXiv:2610.01962 — arXiv (en), https://arxiv.org/abs/2610.01962 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0eea335d545a (Metraipolis archive dc_0eea335d545a, sha256 8f0595acd829de6439df837ae159ba82623df3c779c19cf00ae0424310b10d40)
BibTeX
@misc{dc0eea335d545a,
title = {SIEVE: Selective attention-value Suppression for Vision-Language Models Unlearning — arXiv:2610.01962},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.01962},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0eea335d545a; sha256 8f0595acd829de6439df837ae159ba82623df3c779c19cf00ae0424310b10d40},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01962. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
The ability of vision-language models (VLMs) to associate visual identities with biographical information creates a need for selective unlearning of personally identifiable information (PII) while preserving permitted knowledge about the same individual. This setting is challenging because both sensitive and retained information can share the same visual inputs and intermediate representations. We introduce SIEVE, a simple and effective framework for selective VLM unlearning. SIEVE directly regularizes attention-value representations while also controlling model outputs. SIEVE suppresses attention values for forget examples toward a constant zero, while preserving retain-example representations by matching them to a frozen reference model. These objectives are combined with sequence-level forget and retain supervision, enabling targeted forgetting without largely affecting retained knowledge. Extensive experiments show that SIEVE achieves state-of-the-art performance on unlearning with multiple model-modality settings, while maintaining competitive retained utility. Ablation studies further show that value suppression and negative cross-entropy contribute complementary forgetting signals, while reference-based value matching substantially reduces utility degradation. These results demonstrate that attention values provide an effective intervention point for selective multimodal unlearning when sensitive and retained knowledge are closely related. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CV). Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.01962 — 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 2026-10-06T17:37:36.959Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).