Artificial Intelligence Medical Compendium

Can a vision model learn what words mean from pixels alone? We show that the two largest DINOv3 vision transformers organize words rendered in images by word meaning, despite being pretrained using self-supervision on images only, without paired capt... read more 

Vision-language models (VLMs) exhibit complementary capabilities, yet selecting the most suitable expert for a given image-question query remains challenging due to heterogeneous multimodal inputs and varying expert strengths. Existing routing approa... read more 

Multi-source semantic information fusion enables edge networks to support global inference using observations collected by distributed terminals. However, limited communication and computation resources restrict semantic uploading, leaving the edge s... read more 

Pretrained vision transformers learn structured image representations spanning spatial patch tokens and non-spatial tokens. Recent analyses suggest that non-spatial register tokens encode global scene properties such as illumination, blur, and style.... read more 

Co-speech gesture video generation aims to synthesize realistic, audio-driven portrait videos from a single reference image and a speech signal. Despite advances in visual quality, diffusion-based approaches still suffer from degraded local details a... read more 

Drag-based image editing provides an intuitive interface for spatially controlled image manipulation by allowing users to specify handle and target points. Existing methods have made substantial progress through optimization-based, one-step, and feed... read more 

Image-to-video (I2V) diffusion models can animate a single reference image into high-quality, subject-consistent videos, posing threats of misuse as anyone can turn personal photos into fabricated videos without consent. A promising solution is to pr... read more 

Attention sinks are tokens or positions that consistently receive disproportionate attention mass, often beyond what their semantic content would suggest. Despite extensive study in large language models, their behavior and functional relevance in te... read more 

Regression-based RL for flow models mandates sampling \emph{without} classifier-free guidance (CFG) to align the rectification target with the reference marginal velocity. Extending Advantage-Weighted Rectification (AWR), which dynamically scales thi... read more 

Diffusion feature, i.e., intermediate activations extracted from diffusion model backbones, has emerged as a promising approach for unifying generation and vision understanding (discrimination). Despite the advancement of diffusion backbones into Dif... read more