arXiv Computer Vision
PACE-MoE: Adaptive Domain Sampling and Domain-Specific LoRA for Medical Visual Grounding
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Computer Vision
Oct 27
MedAdaptAgent: An Automatic Harness for Adapting Medical Imaging Models to New Sites
arXiv:2610.34884v1 Announce Type: new Abstract: Reusing pretrained medical-imaging models offers an alternative to training from scratch. A model trained for the same task may perform poorly at a new...
arXiv Computer Vision
Oct 21
AdaptMIL: Domain Adaptation for Multiple Instance Learning in Digital Pathology
arXiv:2610.28007v1 Announce Type: new Abstract: Pathology foundation models (PFMs) support a wide range of histopathology tasks on whole-slide images through multiple instance learning (MIL). Despite...
arXiv AI
Aug 5
CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement
arXiv:2608. 03890v1 Announce Type: cross Abstract: A clinically useful chest X-ray system must go beyond fluent report generation: it should classify findings with tunable decision thresholds, localize them spatially, and derive the anatomical measurements upon which many diagnoses depend.
arXiv Computer Vision
Sep 22
AdaptiveCDM: Source-Free Few-Shot Domain Adaptation for Cell Detection in Microscopic Images
AdaptiveCDM is a modular framework for source‑free few‑shot domain adaptation in cell detection, enabling a pretrained model to adapt to new imaging domains using only a handful of labeled target images and no source data. It combines Resolution‑Aware Augmentation (RAug) to balance scarce, class‑imbalanced samples while preserving cellular morphology, and Category‑Aware Representation Learning (CARL) to strengthen class‑consistent proposals for better localization and classification. Experiments on M5 and Raabin‑WBC datasets show that AdaptiveCDM achieves competitive or superior mAP scores compared to state‑of‑the‑art methods under their respective supervision settings.
arXiv Machine Learning
Aug 31
RegCL: Compact Continual SAM Adaptation for Visual Grounding in Multi-Sensorial Media
RegCL is a non‑replay continual learning framework that adapts the Segment Anything Model (SAM) for visual grounding across evolving multi‑sensorial media domains. It consolidates domain‑specific segmentation knowledge into a single lightweight SAM adapter by incrementally merging LoRA‑style AugModules and preserving compact historical feature statistics. Experiments on five heterogeneous datasets demonstrate that RegCL retains performance while adapting to new domains, outperforming other non‑replay continual learning and merging baselines.
arXiv Computer Vision
Oct 12
Learning Features, Not Layouts: Feature-Axis Attention as a Transfer Prior for Medical Imaging
The paper introduces FAD‑ViT, a vision transformer that decouples spatial and feature axes of global attention, making feature‑axis attention the primary operator. Experiments on 23 medical imaging tasks show that shifting global attention from spatial to feature leaves from‑scratch performance unchanged but significantly boosts transfer learning, achieving up to three times the gain with fewer parameters and lower compute. The study demonstrates that prioritizing feature‑axis attention is a simple, cost‑effective design choice for improving medical image transfer.