{"slug": "exemplar-classical-priors-complement-frozen-features-for-few-shot-microscopy-at", "title": "Exemplar: Classical Priors Complement Frozen Features for Few-Shot Microscopy Segmentation at Native Resolution", "summary": "Researchers introduced Exemplar, a few-shot segmenter that fuses a frozen DINOv3 backbone with classical native-resolution filter responses in a single lightweight head, achieving a mean score of 0.782 across eleven biomedical imaging datasets. The classical bank alone reached 0.693 and frozen features alone 0.672, while Exemplar led in 54 of 55 method-dataset comparisons against five forward-pass few-shot methods, with 52 significant after Holm correction. From a single annotated mask, Exemplar reached 0.703 versus 0.682 for a from-scratch nnU-Net, though nnU-Net overtook it at eight masks but took 16-77x longer to fit.", "body_md": "arXiv:2609.03080v1 Announce Type: new\nAbstract: Segmenting a new biomedical dataset usually means a domain-specific model trained on substantial annotation, or a foundation model steered at inference time. We present Exemplar, a few-shot segmenter that fuses a frozen DINOv3 backbone with a fixed bank of classical native-resolution filter responses in one lightweight head, fitted from the support masks alone. In the few-mask, native-resolution regime, classical priors and frozen self-supervised features are complementary: fused in one head, a single fixed configuration spans eleven biomedical imaging datasets. Under the same head, the classical bank alone reaches 0.693 on the eleven-dataset panel, scored by foreground intersection-over-union or centreline Dice, and the frozen features alone 0.672; the bank leads on seven of the eleven and the features on the rest, and fused they reach 0.782. Against five forward-pass few-shot methods, Exemplar leads in 54 of 55 method-dataset comparisons, 52 of them significant after Holm correction. From a single annotated mask it reaches 0.703 on the same panel, against 0.682 for a from-scratch nnU-Net trained on that same mask. At eight masks nnU-Net overtakes it on the panel mean, chiefly on centreline agreement, but takes 16-77x longer to fit.", "url": "https://wpnews.pro/news/exemplar-classical-priors-complement-frozen-features-for-few-shot-microscopy-at", "canonical_source": "https://arxiv.org/abs/2609.03080", "published_at": "2026-09-04 04:00:00+00:00", "updated_at": "2026-09-04 04:24:22.740174+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision"], "entities": ["Exemplar", "DINOv3", "nnU-Net"], "alternates": {"html": "https://wpnews.pro/news/exemplar-classical-priors-complement-frozen-features-for-few-shot-microscopy-at", "markdown": "https://wpnews.pro/news/exemplar-classical-priors-complement-frozen-features-for-few-shot-microscopy-at.md", "text": "https://wpnews.pro/news/exemplar-classical-priors-complement-frozen-features-for-few-shot-microscopy-at.txt", "jsonld": "https://wpnews.pro/news/exemplar-classical-priors-complement-frozen-features-for-few-shot-microscopy-at.jsonld"}}