Artificial Intelligence Medical Compendium

Explore the latest research on artificial intelligence and machine learning in medicine.

Showing 50,931 to 50,940 of 225,062 articles

Aggregate Models, Not Explanations: Improving Feature Importance Estimation

arXiv
Feature-importance methods show promise in transforming machine learning models from predictive engines into tools for scientific discovery. However, due to data sampling and algorithmic stochasticity, expressive models can be unstable, leading to in... read more 

Light4D: Training-Free Extreme Viewpoint 4D Video Relighting

arXiv
Recent advances in diffusion-based generative models have established a new paradigm for image and video relighting. However, extending these capabilities to 4D relighting remains challenging, due primarily to the scarcity of paired 4D relighting tra... read more 

Efficient Segment Anything with Depth-Aware Fusion and Limited Training Data

arXiv
Segment Anything Models (SAM) achieve impressive universal segmentation performance but require massive datasets (e.g., 11M images) and rely solely on RGB inputs. Recent efficient variants reduce computation but still depend on large-scale training. ... read more 

JEPA-VLA: Video Predictive Embedding is Needed for VLA Models

arXiv
Recent vision-language-action (VLA) models built upon pretrained vision-language models (VLMs) have achieved significant improvements in robotic manipulation. However, current VLAs still suffer from low sample efficiency and limited generalization. T... read more 

Free Lunch for Stabilizing Rectified Flow Inversion

arXiv
Rectified-Flow (RF)-based generative models have recently emerged as strong alternatives to traditional diffusion models, demonstrating state-of-the-art performance across various tasks. By learning a continuous velocity field that transforms simple ... read more 

Free Lunch for Stabilizing Rectified Flow Inversion

arXiv
Rectified-Flow (RF)-based generative models have recently emerged as strong alternatives to traditional diffusion models, demonstrating state-of-the-art performance across various tasks. By learning a continuous velocity field that transforms simple ... read more 

Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception

arXiv
Multimodal Large Language Models (MLLMs) excel at broad visual understanding but still struggle with fine-grained perception, where decisive evidence is small and easily overwhelmed by global context. Recent "Thinking-with-Images" methods alleviate t... read more 

DiffPlace: Street View Generation via Place-Controllable Diffusion Model Enhancing Place Recognition

arXiv
Generative models have advanced significantly in realistic image synthesis, with diffusion models excelling in quality and stability. Recent multi-view diffusion models improve 3D-aware street view generation, but they struggle to produce place-aware... read more 

SynthRAR: Ring Artifacts Reduction in CT with Unrolled Network and Synthetic Data Training

arXiv
Defective and inconsistent responses in CT detectors can cause ring and streak artifacts in the reconstructed images, making them unusable for clinical purposes. In recent years, several ring artifact reduction solutions have been proposed in the ima... read more 

Synthesis of Late Gadolinium Enhancement Images via Implicit Neural Representations for Cardiac Scar Segmentation

arXiv
Late gadolinium enhancement (LGE) imaging is the clinical standard for myocardial scar assessment, but limited annotated datasets hinder the development of automated segmentation methods. We propose a novel framework that synthesises both LGE images ... read more