Live intelligence

Latest AI updates

3 updates
Research

GALA: Gallery-Level Alignment for Real-Life Vision Decoding from EEG

The paper introduces incremental transductive calibration for EEG-based visual decoding, where a decoder is calibrated on unlabeled EEG and a known gallery of candidate images. It proposes GALA, a label-free method using shrinkage whitening, hubness correction, and Sinkhorn-Procrustes alignment with query frequencies estimated from the buffer. On THINGS-EEG-2 with a 200-image gallery, GALA raises top-1 accuracy for a new subject from 28% to 67% over 53 minutes of use. Across 60 evaluation settings, it is reported as the most accurate label-free method, averaging 6.6 points above the best competing method at the end of the stream.

EEG decodingBrain-computer interfacesDomain adaptation
Research

SPSC improves zero-shot cross-subject EEG-to-image reconstruction

Researchers introduce Semantic Preservation and Subspace Calibration (SPSC), a framework for reconstructing images from EEG signals when the target subject provides no labels or image embeddings. SPSC combines Object Mask Contrastive Regularization during training with a Subspace Candidate Gate that performs label-free semantic correction at inference. In leave-one-subject-out experiments across ten subjects on THINGS-EEG, the method improves cross-subject reconstruction from 0.552 to 0.713 on CLIP and from 0.612 to 0.745 on Alex.5. The work targets semantic shift and low signal-to-noise ratio in EEG-based brain-to-image generation.

EEG reconstructionbrain-computer interfacesdiffusion models
Research

BEEMO: Decoupled Behavioral Pretraining Scales Neural Motor Decoding

Researchers introduce BEEMO, a multi-session framework that decouples behavioral and neural representation learning for neural motor decoding. It trains a behavior decoder on abundant unpaired behavioral data and incorporates neural activity via cross-attention when paired recordings are available. Experiments on nonhuman primate datasets show that adding unpaired behavioral data improves decoding performance, especially for transfer to new tasks, and behavior-only pretraining matches models pretrained on larger neural datasets.

neural decodingbehavioral datapretraining