AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

D$^2$-FOSA: Dual-Diffusion Guided EEG-to-Image Reconstruction with Frequency-Oriented Semantic Alignment

Chenglong Yu, Shuai Shen, Xiangsheng Li, Yang Li

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace Mamba with frequency-oriented FOMamba in the EEG encoder (ablation)FSTDE without DDLG · THINGS-EEG 200-waytop-1 accuracy (%): 27.75 -> 31.18Improved
replace Transformer with FOMamba (ablation)FSTDE without DDLG · THINGS-EEG 200-waytop-1 accuracy (%): 25.30 -> 31.18Improved
add full bidirectional DDLG (ablation)FOMamba encoder · THINGS-EEG 200-waytop-1 accuracy (%): 31.18 -> 37.96Improved
add full DDLG (ablation)Mamba encoder · THINGS-EEG 200-waytop-1 accuracy (%): 27.75 -> 30.10Improved
add full DDLG (ablation)Transformer encoder · THINGS-EEG 200-waytop-1 accuracy (%): 25.30 -> 27.30Improved
use bidirectional DDLG instead of only EEG-to-image diffusion (ablation)FOMamba encoder · THINGS-EEG 200-waytop-1 accuracy (%): 35.36 -> 37.96Improved
add DDLG (ablation)FOMamba encoder · THINGS-EEG 50-waytop-5 accuracy (%): 84.58 -> 87.09Improved
bound damping rho_max at 0.995 instead of 1.0 (ablation)FOMamba · THINGS-EEG 200-waytop-1 accuracy (%)Improved
set rho_min around 0.5-0.7 rather than lower or higher (ablation)FOMamba · THINGS-EEG 200-waytop-1 accuracy (%)Improved
D2-FOSA versus MB2C on EEGCVPR40 (5-95 Hz) top-5D2-FOSA · EEGCVPR40 (5-95 Hz filter)top-5 accuracy (%): 99.18 -> 98.75Hurt
D2-FOSA versus VE-SDN on THINGS-EEG intra-subjectD2-FOSA · THINGS-EEG 200-way (intra-subject, avg)top-1 accuracy (%): 37.2 -> 38.0Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy