Tag: Attention Mechanism

Researchers have built a two-stage deep learning system in which a transformer makes an initial electricity demand forecast and an ...

Researchers in India report a hybrid dual-attention pyramidal deep learning model that detects liver tumors on CT scans with 96.52 ...

A new explainable transformer that combines frequency-domain analysis and topology-aware structural modeling achieves 97.86 percent accuracy in classifying eight retinal ...

Researchers in India built a transformer-based deep learning system to detect anomalies in solar-wind power plants, achieving strong reconstruction accuracy ...

Researchers at Sun Yat-sen University have developed an improved convolutional prototype network that classifies known environmental sounds and rejects unknown ...

A new difficulty-aware Transformer framework jointly optimized by particle swarm intelligence improves sports score trend prediction while attributing forecasts to ...

A new deep learning model combines diffusion-style state propagation with Mamba-based sequence modeling and window attention to grade ulcerative colitis ...

A new detection framework combines hybrid attention, a slim dynamic head, and a novel loss function to spot crop diseases ...

A new cloud-edge AI framework called SMWPF combines a lightweight edge predictor with a Qwen-based cloud model to cut wind ...

Researchers have developed a wavelet-transform-based pedestrian detection model that improves accuracy on crowded benchmarks while cutting parameters by nearly 40 ...

A hybrid CNN–BiLSTM network with additive attention classifies brain MRI into four Alzheimer's disease stages with 95.94% slice-level accuracy, while ...

A new deep learning architecture called FrDPalmNet combines fractional calculus, dual attention, and densely connected blocks to push palmprint-based biometric ...

Researchers at Central South University have created a hybrid deep learning model that lets mobile agricultural robots accurately forecast pollutant ...

A new parallel Transformer-CNN architecture called TCNNISTD achieves state-of-the-art accuracy and real-time speed in detecting infrared targets smaller than ten ...

A two-stage imaging and deep learning framework detects and grades infestations of four greenhouse pests, distilling 248 spectral bands into ...

Researchers have integrated transformer attention mechanisms into learned primal-dual PET reconstruction, achieving cleaner images from noisy low-count scanner data.

A new self-supervised framework combining joint-embedding prediction and bidirectional Mamba learns diagnostic ECG representations without labeled data, outperforming prior baselines ...

A spatiotemporal attention network that reads fNIRS blood-oxygen signals while patients listen to music classifies levels of consciousness with 88.72 ...

A new machine learning model called DBFP combines LSTM networks, attention mechanisms, and non-linear regularization to predict which software features ...

A new FCN-GRU-Attention deep learning model with a confidence correction mechanism identifies overlapping composite power quality disturbances with over 99 ...