{"ID":5443743,"CreatedAt":"2026-07-01T02:07:11.383974684Z","UpdatedAt":"2026-07-03T13:34:17.08154537Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.31664","arxiv_id":"2606.31664","title":"Sparsity-Inducing Divergence Losses for Biometric Verification","abstract":"Performance in face and speaker verification is largely driven by margin-penalty softmax losses such as CosFace and ArcFace. Recently introduced $α$-divergence loss functions offer a compelling alternative, particularly due to their ability to induce sparse solutions (when $α\u003e1$). However, standard geometric margins are designed for the softmax function and do not naturally extend to this generalized probabilistic framework. In this paper we propose Q-Margin, a novel $α$-divergence loss that introduces a principled probabilistic margin. Unlike conventional methods that apply geometric penalties to the logits (unnormalized log-likelihoods), Q-Margin encodes the margin penalty directly into the reference measure (prior probabilities). This formulation naturally encourages discriminative embeddings while preserving the beneficial sparsity properties of the $α$-divergence. We demonstrate that Q-Margin achieves competitive or superior performance on the challenging IJB-B and IJB-C face verification benchmarks and similarly strong results in speaker verification on VoxCeleb. Crucially, against ArcFace and CosFace baselines trained under an identical recipe, Q-Margin consistently improves at low False Acceptance Rates (FARs), a capability critical for practical high-security applications. Finally, the extreme sparsity of the Q-Margin posteriors enables exact and memory-efficient training, offering a scalable solution for datasets with millions of identities.","short_abstract":"Performance in face and speaker verification is largely driven by margin-penalty softmax losses such as CosFace and ArcFace. Recently introduced $α$-divergence loss functions offer a compelling alternative, particularly due to their ability to induce sparse solutions (when $α\u003e1$). However, standard geometric margins ar...","url_abs":"https://arxiv.org/abs/2606.31664","url_pdf":"https://arxiv.org/pdf/2606.31664v1","authors":"[\"Dimitrios Koutsianos\",\"Ladislav Mošner\",\"Yannis Panagakis\",\"Themos Stafylakis\"]","published":"2026-06-30T13:42:41Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[]","has_code":false}