Feature from recent image foundation models (DINOv2) are useful for vision tasks (segmentation, object localization) with little or no human input. Once upsampled, they can be used for weakly supervised micrograph segmentation, achieving strong results when compared to classical features (blurs, edge detection) across a range of material systems.
Ronan Docherty +2 more
wiley +1 more source
Middle-schoolers' reading and lexical-semantic processing depth in response to digital and print media: An N400 study. [PDF]
Froud K, Levinson L, Maddox C, Smith P.
europepmc +1 more source
Terrestrial Cyborg Insects for Real‐Life Applications
This article reviews the development of terrestrial cyborg insects from their emergence in 1997 to mid‐2025, examining three key aspects: locomotion control methods, associated challenges with proposed solutions, and practical applications. Framing these biohybrid systems as insect‐scale mobile robots, the review provides foundational insights for new ...
Hai Nhan Le +10 more
wiley +1 more source
This study provides an introduction to Bayesian optimisation targeted for experimentalists. It explains core concepts, surrogate modelling, and acquisition strategies, and addresses common real‐world challenges such as noise, constraints, mixed variables, scalability, and automation.
Chuan He +2 more
wiley +1 more source
White matter network underlying semantic processing: evidence from stroke patients. [PDF]
Xiao X +7 more
europepmc +1 more source
Human‐Machine Mutual Trust Based Shared Control Framework for Intelligent Vehicles
This work introduces a bidirectional‐trust‐driven shared control framework for human‐machine co‐driving. The method models human‐to‐machine trust from intention discrepancies and Bayesian skill assessment, and machine‐to‐human trust from integrated ability evaluation.
Zhishuai Yin +4 more
wiley +1 more source
Semantic processing in older adults is associated with distributed neural activation which varies by association and abstractness of words. [PDF]
Garcia A +6 more
europepmc +1 more source
KDLM: Lightweight Brain Tumor Segmentation via Knowledge Distillation
A lightweight student network is designed, which is based on multiscale and multilevel feature fusion and combined with the residual channel attention mechanism to achieve efficient feature extraction and fusion with very few parameters. A dual‐teacher collaborative knowledge distillation framework is proposed.
Baotian Li +4 more
wiley +1 more source
Scalable High‐Force Zipping Electrostatic Actuators
An electrostatic actuator architecture is presented that combines high energy density with forces and strokes suited for wearable soft robotics. By folding patterned thin films into a scalable honeycomb of zipping units, performance becomes tunable through series–parallel design, achieving over 40 N and 28 J/kg.
Fabio Caruso +2 more
wiley +1 more source
An implemented predictive coding model of lexico-semantic processing explains the dynamics of univariate and multivariate activity within the left ventromedial temporal lobe during reading comprehension. [PDF]
Wang L +4 more
europepmc +1 more source

