Decoding visual neural representations by multimodal with dynamic balancing

Kaili Sun

doi:10.1016/j.eswa.2025.129382

Decoding visual neural representations by multimodal with dynamic balancing

2025-08-18

0

PRE

AI

DOI：10.1016/j.eswa.2025.129382

Original

Original request for help

Save

Abstract

En 中文

• A novel HMAVD fuses EEG, image, and text to boost visual decoding. • Text bridges EEG and image, stabilizing high-dim features for alignment. • Dynamic MCDB and SPR balance modality gradients, boosting robustness.

Keywords:

HMAVD

EEG

image

text

multimodal fusion

Decoding visual neural representations by multimodal with dynamic balancing

Abstract

Journal

Organization