# zero-shot classification

3 related articles

Subscribe to updates

arXiv Oct 24, 2026
Medical image pretraining stops reading only reports: DICOM acquisition fields enter contrastive learning, lifting zero-shot CT macro-AUROC by 1.41 points
Medical vision-language pretraining learns from paired images and radiology reports, but scanner and acquisition-protocol differences can impair transfer. Meta-CLIP uses structured acquisition information in DICOM headers as a training-time prior: a kernel over numerical and categorical fields defines a bounded pairwise margin that reweights cross-modal negatives during pretraining, while inference keeps the standard dual-encoder architecture and needs no metadata. On CT-RATE it improves macro-AUROC over CT-CLIP by 1.41 points, with a 95% paired bootstrap interval of 0.96 to 1.85 points, and reaches the highest macro-AUROC on four of five chest radiography benchmarks.
Medical vision-language pretraining learns from paired images and radiology reports, but scanner and acquisition-protocol differences can impair transfer. Meta-CLIP uses structured acquisition information in DICOM headers as a training-time prior: a kernel over numerical and categorical fields defines a bounded pairwise margin that reweights cross-modal negatives during pretraining, while inference keeps the standard dual-encoder architecture and needs no metadata. On CT-RATE it improves macro-AUROC over CT-CLIP by 1.41 points, with a 95% paired bootstrap interval of 0.96 to 1.85 points, and reaches the highest macro-AUROC on four of five chest radiography benchmarks.
Haozhe Luo（First author）
Meta-CLIP metadata-conditioned contrastive pretraining zero-shot classification
arXiv Sep 30, 2026
SentZero pairs abstract-level sentence mapping with patch-level false-negative alignment to beat prior multi-task zero-shot methods on chest X-ray classification and grounding after MIMIC-CXR pretraining
SentZero is a sentence-centric vision-language pretraining framework for chest X-ray: it uses an LLM to extract phrases from radiology reports and map them to concise topic-presence sentences that expand positive-pair diversity, adds an auxiliary loss that attracts the highest-similarity image patches toward clinical sentences recurring across studies to mitigate false negatives, and applies sentence-conditioned residual modulation to visual embeddings; after pretraining on MIMIC-CXR it outperforms prior multi-task zero-shot methods on most metrics for zero-shot multi-label classification on Open-I, ChestXray14, PadChest, ChestXDet10 and CheXpert and for zero-shot grounding on ChestXDet10 and MS-CXR.
SentZero is a sentence-centric vision-language pretraining framework for chest X-ray: it uses an LLM to extract phrases from radiology reports and map them to concise topic-presence sentences that expand positive-pair diversity, adds an auxiliary loss that attracts the highest-similarity image patches toward clinical sentences recurring across studies to mitigate false negatives, and applies sentence-conditioned residual modulation to visual embeddings; after pretraining on MIMIC-CXR it outperforms prior multi-task zero-shot methods on most metrics for zero-shot multi-label classification on Open-I, ChestXray14, PadChest, ChestXDet10 and CheXpert and for zero-shot grounding on ChestXDet10 and MS-CXR.
Hangyul Yoon（First author）
SentZero Vision-language (VL) pretraining Contrastive Learning zero-shot classification Visual Grounding
arXiv Sep 14, 2026
Just Leaf It: Accelerating Diffusion Classifiers with Hierarchical Class Pruning
The work proposes a Hierarchical Diffusion Classifier (HDC) that exploits parent-child label hierarchies: in a pruning stage it traverses the label tree level by level, estimates epsilon-prediction errors with fewer Monte Carlo samples, and keeps only the lowest-error nodes, then runs classical diffusion classification on the surviving leaf nodes, achieving roughly 60% inference speed-up on ImageNet-1K with Stable Diffusion 2.0 (1600s to 650s) or raising per-class accuracy from 64.90% to 65.16% at nearly unchanged runtime.
The work proposes a Hierarchical Diffusion Classifier (HDC) that exploits parent-child label hierarchies: in a pruning stage it traverses the label tree level by level, estimates epsilon-prediction errors with fewer Monte Carlo samples, and keeps only the lowest-error nodes, then runs classical diffusion classification on the surviving leaf nodes, achieving roughly 60% inference speed-up on ImageNet-1K with Stable Diffusion 2.0 (1600s to 650s) or raising per-class accuracy from 64.90% to 65.16% at nearly unchanged runtime.
German Research Centre for Artificial Intelligence University of Koblenz and Landau Rheinland-Pfälzische Technische Universität Kaiserslautern-Landau University of Kaiserslautern
Arundhati S. Shanbhag（First author）Brian B. Moser（Corresponding author）
Hierarchical Diffusion Classifier Hierarchical Class Pruning zero-shot classification Image Classification Diffusion Model

You’re all caught up

Medical image pretraining stops reading only reports: DICOM acquisition fields enter contrastive learning, lifting zero-shot CT macro-AUROC by 1.41 points

SentZero pairs abstract-level sentence mapping with patch-level false-negative alignment to beat prior multi-task zero-shot methods on chest X-ray classification and grounding after MIMIC-CXR pretraining

Just Leaf It: Accelerating Diffusion Classifiers with Hierarchical Class Pruning