ConQuer augments global CLIP alignment with independent per-concept contrastive losses on anatomical regions extracted from reports, producing Jolia which outperforms CLIP baselines on classification, report generation, and transfer.
Curia-2: Scaling self-supervised learning for radiology foundation models.arXiv preprint arXiv:2604.01987, 2026
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2representative citing papers
Benchmark finds segmentation dominates volume and stage tasks while classifier choice dominates survival, histology, and age prediction, recommending Curia with tumor segmentation and CatBoost as a safe default.
citing papers explorer
-
Jolia: Concept-Level Vision-Language Alignment for 3D CT Contrastive Learning
ConQuer augments global CLIP alignment with independent per-concept contrastive losses on anatomical regions extracted from reports, producing Jolia which outperforms CLIP baselines on classification, report generation, and transfer.
-
Foundation Models vs. Radiomics for Lung Computed Tomography: A Benchmark of Feature Extractors, Classification Heads, and Segmentation Choices
Benchmark finds segmentation dominates volume and stage tasks while classifier choice dominates survival, histology, and age prediction, recommending Curia with tumor segmentation and CatBoost as a safe default.