Foundation models for medical imaging are typically pretrained on increasingly large datasets, following a "scale-at-all-costs" paradigm. However, this strategy faces two critical challenges: large-scale medical datasets often contain substantial red... read more
As Large Multimodal Models (LMMs) scale up and reinforcement learning (RL) methods mature, LMMs have made notable progress in complex reasoning and decision making. Yet training still relies on static data and fixed recipes, making it difficult to di... read more
This paper focuses on enhancing the grasping precision and generalization of manipulation policies learned via imitation learning. Diffusion-based policy learning methods have recently become the mainstream approach for robotic manipulation tasks. As... read more
In this article, we provide an axiomatic characterization of feature attribution for multi-output predictors within the Shapley framework. While SHAP explanations are routinely computed independently for each output coordinate, the theoretical necess... read more
Human intelligence naturally intertwines omni-modal perception -- spanning vision, audio, and language -- with complex reasoning and tool usage to interact with the world. However, current multi-modal LLMs are primarily confined to bi-modal interacti... read more
This study investigates a data-driven machine learning approach to predict membrane fouling in critically ill patients undergoing Continuous Renal Replacement Therapy (CRRT). Using time-series data from an ICU, 16 clinically selected features were id... read more
A clinically actionable Cardiac Digital Twin (CDT) should reconstruct individualised cardiac anatomy and physiology, update its internal state from multimodal signals, and enable a broad range of downstream simulations beyond isolated tasks. However,... read more
Although deep learning has significantly advanced the perception capabilities of intelligent transportation systems, railway applications continue to suffer from a scarcity of high-quality, annotated data for safety-critical tasks like obstacle detec... read more
We introduce Manifold Sobolev Informed Neural Optimization (MSINO), a curvature aware training framework for neural networks defined on Riemannian manifolds. The method replaces standard Euclidean derivative supervision with a covariant Sobolev loss ... read more
Parameter-efficient fine-tuning has demonstrated promising results across various visual adaptation tasks, such as classification and segmentation. Typically, prompt tuning techniques have harnessed knowledge from a single pre-trained model, whether ... read more
Join thousands of healthcare professionals staying informed about the latest AI breakthroughs in medicine. Get curated insights delivered to your inbox.