Tag: state-space models

A controlled benchmark study finds that 4-bit quantization of a compact Transformer model delivers the strongest overall efficiency gains, while ...

A new 3D Mamba-based model segments brain tumors with consistent accuracy gains and a 15x efficiency improvement over existing state-of-the-art ...

An open benchmark on PRISMA hyperspectral imagery over Egypt's Wadi El-Gemal area found the S2Mamba state-space model achieved the highest ...

A new spatiotemporal model called RoST-Mamba combines adaptive graph learning, frequency-domain filtering and graph-constrained state-space modeling to keep traffic forecasts ...

Researchers have adapted Meta's Segment Anything Model for medical imaging by combining low-rank fine-tuning, Mamba-based adapters, and a prompt-free decoder, ...

Researchers at VIT-AP University have developed TF-Mamba-DPHNet, a compact Mamba-based AI framework that removes noise from speech while preserving fine ...

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

Researchers in China have developed a state-space AI framework that combines spectral attention and Laplace-domain transient decoding to forecast abrupt, ...

A new end-to-end AI detector called SFM-DETR combines Mamba state space modeling with frequency-domain analysis to identify maize seedlings and ...

A Mamba-CNN deep learning framework analyzing brain scans from 1,175 participants detects autism with 87.49% accuracy and reveals novel gray ...