{"slug": "beyond-points-spherical-distributional-part-prototypes-for-interpretable", "title": "Beyond Points: Spherical Distributional Part Prototypes for Interpretable Classification", "summary": "Researchers propose vMFProto, a distributional part-prototype framework that models each class as a mixture of von Mises-Fisher components on the hypersphere for interpretable classification. The method achieves state-of-the-art explanation quality with competitive accuracy on CUB-200-2011, Stanford Dogs, and Stanford Cars datasets.", "body_md": "arXiv:2606.27582v1 Announce Type: new\nAbstract: Prototype-based neural networks aim to provide intrinsic interpretability by grounding predictions in a small set of part prototypes. However, modern vision backbones typically operate in normalized, directional embedding spaces where each semantic part exhibits substantial intra-class variability. As a result, point prototypes often become redundant or unstable, hurting both explanation quality and robustness. We propose vMFProto, a distributional part-prototype framework that models each class as a mixture of von Mises-Fisher components on the hypersphere. Each prototype learns its own concentration, capturing part-specific variability, and we use entropic optimal transport (OT) to obtain structured patch-to-prototype assignments. A two-stage training schedule performs OT-driven prototype discovery followed by end-to-end refinement with patch-level distillation and distribution-aware diversity regularization. Experiments on CUB-200-2011, Stanford Dogs, and Stanford Cars with frozen DINO backbones show that vMFProto achieves state-of-the-art explanation quality (consistency, stability, and distinctiveness) with competitive accuracy. Qualitative results confirm that vMFProto yields localized, non-redundant part evidence.", "url": "https://wpnews.pro/news/beyond-points-spherical-distributional-part-prototypes-for-interpretable", "canonical_source": "https://arxiv.org/abs/2606.27582", "published_at": "2026-06-29 04:00:00+00:00", "updated_at": "2026-06-29 04:03:19.607455+00:00", "lang": "en", "topics": ["machine-learning", "computer-vision", "neural-networks"], "entities": ["vMFProto", "CUB-200-2011", "Stanford Dogs", "Stanford Cars", "DINO"], "alternates": {"html": "https://wpnews.pro/news/beyond-points-spherical-distributional-part-prototypes-for-interpretable", "markdown": "https://wpnews.pro/news/beyond-points-spherical-distributional-part-prototypes-for-interpretable.md", "text": "https://wpnews.pro/news/beyond-points-spherical-distributional-part-prototypes-for-interpretable.txt", "jsonld": "https://wpnews.pro/news/beyond-points-spherical-distributional-part-prototypes-for-interpretable.jsonld"}}