The Flow has not summarised this story yet — read it at Hugging Face Trending Papers.
arXiv:2609.10333v1 Announce Type: new Abstract: Uncertainty estimation for medical vision--language models (VLMs) using conformal prediction has gained increasing attention due to its distribution-fr...
Uncertainty estimation for medical vision--language models (VLMs) using conformal prediction has gained increasing attention due to its distribution-free coverage guarantees. However, standard conform...
The paper investigates how pre‑training strategy, dataset size, and domain affect uncertainty estimation in vision medical foundation models. It compares point‑prediction calibration with conformal (region) prediction across retinal, histopathological, and chest X‑ray models, finding that domain‑specific, self‑supervised pre‑training yields better calibration and more efficient conformal sets. The study shows that standard recalibration alone cannot fully reconcile uncertainty differences between models trained on different data sources.
arXiv:2608. 00632v1 Announce Type: new Abstract: Modern machine learning pipelines increasingly rely on reusing pretrained and foundation models across downstream tasks.
arXiv:2601. 11670v3 Announce Type: replace-cross Abstract: Pseudo-label selection in semi-supervised learning is commonly driven by maximum-confidence thresholds, yet confidence alone can be unreliable under model overconfidence and class imbalance.
arXiv:2608. 12100v1 Announce Type: cross Abstract: In high-stakes applications, reliable confidence estimates are as important as the predictions themselves.