2026 |
Estrada, Jose Edgar Hernandez Cancino; Diaz, Mauro; Emeršič, Žiga; Štruc, Vitomir; Peer, Peter; Tomašević, Darian PIU: Proximity-guided Identity Unlearning in ID-Conditioned Diffusion Models Proceedings Article In: Proceedings of the IEEE International Joint Conference on Biometrics (IJCB 2026), pp. 1–10, 2026. Abstract | Links | BibTeX | Tags: face generation, generative AI, unlearning @inproceedings{Unlearning2026,Identity-conditioned diffusion models enable high-quality and identity-consistent face generation, but they also raise severe privacy concerns, as models may continue to synthesize individuals despite their right to be forgotten. While machine unlearning has been extensively studied for concept and data removal, identity unlearning remains largely unexplored, particularly in models conditioned directly on identity embeddings rather than text prompts. In this work, we study identity unlearning in Arc2Face, a state-of-the-art identity-conditioned latent diffusion model for face generation, and introduce textbf{P}roximity-guided textbf{I}dentity textbf{U}nlearning (textbf{PIU}), an anchor-guided framework for identity unlearning. Specifically, we formulate identity removal as an identity replacement objective that reassigns the source identity to a selected anchor identity in the learned identity space, and we complement it with a proximity-based anchor selection strategy motivated by the geometry of ArcFace representations. We further show that effective unlearning can be achieved through localized fine-tuning of a small subset of identity-sensitive cross-attention layers. Experiments across multiple target identities show that our framework effectively suppresses generation of the target identity while preserving realism and identity consistency for retained identities, as validated by improved performance on unlearning and image-quality metrics, together with qualitative evaluation. The source code for PIU is publicly available at url{https://github.com/edgarcancinoe/piu_unlearning} |