Tag: Attention Mechanisms

A systematic survey evaluates modern deep learning architectures and benchmark datasets for traffic volume prediction across recent transport studies.

A hierarchical attention-based reinforcement learning framework improved slice admission rates by up to 20 percent and cut resource overhead by ...

A physics-guided deep learning model trained on five years of monitoring data from the 1,650-metre Xihoumen Bridge predicts wind-induced buffeting ...

A PRISMA-aligned systematic review of 19 GAN-based text-to-image models finds encoder capacity and cross-modal alignment drive generalisation, while a controllability–diversity ...

A new study reports three neural modules that cut computational cost by 26.7 percent while improving accuracy benchmarks for real-time ...

A systematic review of 43 studies finds that attention-based multimodal deep learning consistently outperforms single-modality approaches across lung cancer diagnostic ...

Researchers introduce an attention-guided cross-representation distillation framework that utilizes MRI scans to achieve high accuracy in identifying subtle stages of ...

A hybrid model combining bidirectional deep learning, attention and fuzzy reasoning reported 98.8% accuracy on a 2,500-observation macro-financial policy dataset ...

AMA-YOLOv8s reached 93.5% mAP@0.5 in a dataset-based evaluation of photovoltaic defect detection.

A 2025 Shandong field study found that attention-guided feature selection enabled a UAV random forest to estimate maize maturity with ...

A computational study reports that ADGB-ONet achieved a 2.88% mean relative error when predicting scattered pressure fields, outperforming the compared ...

A systematic review of 168 primary studies found that 57.14% did not examine the performance–faithfulness relationship, 41.07% reported synergy and ...

A hybrid AI recommendation system that combines specialized expert routing with neural modeling of reading behavior reached 81.23% recall for ...

Researchers at Xinjiang University developed BGAFS-UNet, a boundary-aware deep learning architecture that improves retinal vessel segmentation fidelity while openly documenting ...

A new RGB-depth salient object detection framework combines multi-level attention-based feature fusion with a WebP compression and denoising defense that ...

Researchers at Dalian Maritime University have developed ESTGCN, a graph neural network that dynamically rewires regional traffic relationships to deliver ...

Researchers have built a 10.97-million-parameter neural network that classifies brain tumors on MRI with 99.2 percent accuracy while quantifying its ...

Researchers at Durham University have developed a dual-stream neural network that applies parallel attention to sensor modalities and time steps, ...

Researchers have developed CL3AN-GNN, a graph neural network that borrows curriculum learning from education to progressively disentangle features and boost ...

A new 3D convolutional neural network called COV-DSNet stages COVID-19 patients as progressing or remitting from chest CT scans and ...