AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

EEGiT: Teaching Vision Transformers to Understand the EEG signal

Jiahao Zhou, Chenghao Xu, Wei Wang, Erkun Yang, Cheng Deng

CVPR 2026 · Architectures and efficiency › Attention and transformers · approach: Token merging and masked pretraining

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
add both pretrained ViT weights and EEG patch representation (ablation)ViT EEG encoder · THINGS-EEG intra-subjecttop-1 accuracy: 41.8 -> 70.4Improved
add both pretrained ViT weights and EEG patch representation (ablation)ViT EEG encoder · THINGS-EEG inter-subjecttop-1 accuracy: 17.4 -> 24.0Improved
remove one of the two components (second ablation row) (ablation)EEGiT · THINGS-EEG intra-subjecttop-1 accuracy: 70.4 -> 63.6Hurt
remove the other component (third ablation row) (ablation)EEGiT · THINGS-EEG intra-subjecttop-1 accuracy: 70.4 -> 54.0Hurt
add both pretrained ViT and EEG patch (ablation)ViT EEG encoder · EEG-3D color classificationtop-1 accuracy: 34.26 -> 41.32Improved
add both pretrained ViT and EEG patch (ablation)ViT EEG encoder · EEG-3D object classificationtop-1 accuracy: 4.28 -> 5.95Improved
initialize with pretrained ViT weights rather than random (ablation)EEGiT · THINGS-EEG retrievalconvergence speed: about 10 epochs vs about 20 epochsImproved
keep only occipital channels instead of all channels (ablation)EEGiT · THINGS-EEG retrievaltop-1/top-5 accuracyNo clear effect
remove occipital region signals (ablation)EEGiT · THINGS-EEG retrievaltop-1/top-5 accuracyHurt
EEGiT versus Neuro-3D on object classificationEEGiT · EEG-3D object typetop-1 accuracy: 5.91 -> 5.95No clear effect

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