Results 131 to 140 of about 987,911 (296)

Closing the Empirical Loop: Autonomous AI Agents Conduct End‐to‐end Research With Human Participants

Advanced Science, EarlyView.
A multi‐agent AI system autonomously executes the complete scientific workflow, from hypothesis to manuscript, across three psychological studies involving 288 participants. The system designs experiments, collects real world data, develops analysis pipelines, and writes manuscripts with theoretical rigor comparable to experienced researchers.
Gabrielle Wehr, Reuben Rideaux, Amaya J. Fox, David R. Lightfoot, Jason Tangen, Jason B. Mattingley, Shane E. Ehrhardt   +6 more
wiley   +1 more source

A Decoupled Segmentation-Classification Strategy Based on Semantic-SAM for Precise Semantic Segmentation in Coal Mine Areas

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
To address complex semantic segmentation in coal mine areas, this study proposes the SAM-SEF (SAM-based Semantic Enhancement Framework). It integrates Semantic-SAM’s zero-shot segmentation capability with specialized deep learning models through a
Libing Wang, Jihong Dong, Lei Wang, Feng Liu, Qingke Wen, Bekir Genc   +5 more
doaj   +1 more source

Pattern‐Aware Intelligence Enables Nondestructive, Rapid Quantification of High‐Aspect‐Ratio Silicon Etching

Advanced Science, EarlyView.
Pattern‐dependent etching is converted into a physical prior for intelligent reconstruction of high‐aspect‐ratio silicon structures. Combining YOLO‐Pose feature extraction with a topography network, the framework retrieves depth, sidewall angle, and scallop texture from minimal destructive observations, enabling accurate cross‐scale metrology and near ...
Shuyan He, Lang Chen, Bo Wen, Yufeng Jin, Wei Wang   +4 more
wiley   +1 more source

SEMANTIC SEGMENTATION OF MICROSCOPIC BLOOD IMAGE DATA USING SELF-TRAINING TO AUGMENT SMALL TRAINING SETS AND ITS APPLICATION FOR COUNTING CELLS

, 2019
Semantic segmentation is a computer vision task of assigning a label describing the content to each pixel in an image. There has been a lot of progress in this area using deep neural networks with an encoder-decoder structure.
Luo, Junliang
core  

CoSP: Reconfigurable Metamaterial Inverse Design via Contrastive Pretrained Large Language Model

Advanced Science, EarlyView.
In this work, CoSP (contrastive multi‐state pretrain), an intelligent inverse design method for reconfigurable metamaterials based on a contrastive pretrained large language model, is proposed. Numerical experiments demonstrate that CoSP can design reconfigurable metamaterial structures for multi‐state, multi‐band optical responses, showing great ...
Shujie Yang, Yuqi Zhang, Xuzhe Zhao, Yansong Tang, Kaichen Dong   +4 more
wiley   +1 more source

Brain-Inspired Synergistic Adversarial Framework for Style Transfer-Guided Semantic Segmentation in Cross-Domain Remote Sensing Imagery

Remote Sensing
Domain shifts pose significant challenges for cross-domain semantic segmentation in high-resolution remote sensing imagery. Inspired by the cognitive mechanisms of the human brain, we propose a Brain-Inspired Style Transfer and Semantic Segmentation ...
Xinyao Wang, Haitao Wang, Yuqian Jing, Xiaodong Li, Xianming Yang   +4 more
doaj   +1 more source

Decoupling Continual Semantic Segmentation

Proceedings of the AAAI Conference on Artificial Intelligence
Continual Semantic Segmentation (CSS) requires learning new classes without forgetting previously acquired knowledge, addressing the fundamental challenge of catastrophic forgetting in dense prediction tasks. However, existing CSS methods typically employ single-stage encoder-decoder architectures where segmentation masks and class labels are tightly ...
Yifu Guo, Yuquan Lu, Wentao Zhang 0005, Zishan Xu, Dexia Chen, Siyu Zhang, Yizhe Zhang 0001, Ruixuan Wang   +7 more
openaire   +4 more sources

SemanticST: A Scalable Multi‐Contextual Graph Learning Framework for Uncovering Spatial Niches and Robust Multi‐Sample Integration in Spatial Transcriptomics

Advanced Science, EarlyView.
Technical limitations often let dominant signals overshadow rare cell types and fine‐grained heterogeneity in spatial transcriptomics. SemanticST, a graph neural network using multi‐semantic graph fusion and a novel min‐cut loss, recovers these subtle patterns.
Roxana Zahedi, Ahmadreza Argha, Nona Farbehi, Ivan Bakhshayeshi, Thantrira Porntaveetus, Youqiong Ye, Nigel H. Lovell, Hamid Alinejad‐Rokny   +7 more
wiley   +1 more source

Learning Work Function via Implicit Reasoning on Electrostatic Potential Landscapes

Advanced Science, EarlyView.
StructPot‐CLR establishes a cross‐modal contrastive learning framework that aligns the crystal structures of 2D materials with plane‐averaged electrostatic potential landscapes for physically informed work‐function prediction. The model achieves an MAE of 0.265 eV and an R2 of 0.902 on the held‐out test set while accurately preserving key morphological
Haoyu Wan, Yue Wu, Tianhao Su, Deng Pan
wiley   +1 more source

A Site‐Aware Representation Learning Framework For Unified Molecular Interaction Modeling and Generative Design

Advanced Science, EarlyView.
MolDBG is a site‐aware, sequence‐only framework that unifies drug‐target affinity prediction, binding‐site identification, and affinity‐conditioned molecular generation for structured proteins. Guided by multi‐task binding‐site supervision, it aligns interaction‐critical residues before learning drug‐target representations and simultaneously infers ...
Gang Luo, Qianqian Zhang, Chenhao Wang, Zhiheng Yi, Alex Jinpeng Wang, Jing Tang, Min Li   +6 more
wiley   +1 more source