eeg

BrainBench is a new unified benchmark for evaluating large language models on comprehensive, instruction-conditioned EEG understanding, covering 17 datasets, 172 tasks, and over 4K real-data instances. The paper evaluates 13 LLMs across two execution paradigms, showing that EEG competence varies by model and operationalization.

This paper proposes a multi-representation deep learning framework combining CNNs, LSTMs, and recurrence plot analysis to characterize alpha and gamma EEG biomarkers in Fragile X Syndrome, showing improved classification over single-modality baselines.

A comprehensive review of technological advances in detecting and managing cognitive impairment in older adults, covering EEG, neuroimaging, blood-based biomarkers, digital tools, and AI/ML integration, with a focus on challenges like external validation and equitable deployment.

This paper examines how evaluation protocols affect reported accuracy in EEG emotion recognition, using a DGCNN on SEED and SEED-IV datasets. It demonstrates that subject-dependent, subject-disjoint, and cross-session evaluations answer different questions, and that checkpoint selection and test-set reuse can inflate accuracy.

Introduces ZUNA1.1, a 380M-parameter diffusion autoencoder for flexible EEG signal reconstruction, capable of handling variable-length sequences, arbitrary channels, and temporal intervals, while outperforming standard interpolation methods. The model is released open source under the Apache 2.0 license.

This paper proposes BridgeMIL, a two-stage framework for EEG-based disease diagnosis that decouples instance representation learning from subject-level supervision using multiple instance learning. It achieves state-of-the-art accuracy on three EEG disease datasets, outperforming strong baselines.

CogEEGAgent is an LLM-based agent for autonomous cognitive EEG analysis that uses a grounded execution and selection-aware verification framework to ensure reliable analysis choices and block false positives from adaptive search.

This paper presents a large-scale study comparing Bayesian complete-pooling models to frequentist baselines for cross-subject motor imagery EEG classification, finding that Bayesian methods improve reliability modestly but at a higher computational cost, with limited practical benefit.

This paper proposes a lightweight CNN classifier that uses Fourier-Bessel Series Expansion with Euclidean Distance (FBSE-ED) to represent EEG signals as images, achieving 93.60% accuracy in predicting the outcome of rTMS depression therapy, outperforming both EEG-specific and pretrained deep learning models.

This study uses EEG to investigate how word predictability influences N400 brain responses across lexical categories, showing that content words exhibit greater N400 differences than function words, and decoding techniques outperform traditional ERP analysis.

This paper investigates whether language models' next-word prediction aligns with human cognitive processing by analyzing EEG signals and event-related potentials, finding that only surprisal correlates with human brain responses, especially for open-class words.

This paper presents a comprehensive comparison of deep learning architectures, including Vision Transformers and Graph Attention Networks, for automated sleep apnea detection from multichannel EEG signals, achieving a best test AUC of 0.750 using a vision transformer trained on topological data analysis features.

Gidi Littwin, co-inventor of Apple's FaceID and Vision Pro, is building a frontier AI model for decoding brain electrical activity to diagnose cognitive disorders. His startup Hemispheric raised $52 million after collecting brain data from 100,000 volunteers, and plans to submit its PTSD diagnostic tool for FDA approval in early 2027.

This paper introduces a multi-feature fusion framework for semantic reconstruction from non-invasive brain recordings, combining static lexical (Word2Vec) and dynamic contextual (GPT) representations via cross-attention, achieving state-of-the-art performance in brain-to-text decoding.

This paper introduces the RG-Flow Transformer, a model with a renormalization-group inductive bias for analyzing scarce EEG data. It benchmarks against a vanilla transformer on sleep staging from the Sleep-EDF dataset, finding no accuracy advantage but better interpretability through recovery of the spectral exponent.

The paper introduces a graph-regularized deep learning framework for EEG-based emotion recognition that incorporates psychologically-grounded emotion topology into the training objective, achieving up to +5.42% accuracy and 39% reduction in psychologically implausible misclassifications on SEED datasets.

This study examines how infants' brain responses and spontaneous movements to music develop over the first year, finding that auditory encoding of music emerges early but coordinated movement patterns appear only by 12 months.

Introduces STST-JEPA, a self-supervised transformer for EEG that predicts masked-token representations, pretrained on 47,703 sessions for brain age regression across ages 5–81.

Proposes Stacked LoRA, a framework that decouples subject-invariant and subject-specific knowledge for adapting EEG foundation models to motor imagery decoding, achieving improved accuracy across multiple benchmarks.

This paper presents a granularity-aware EEG feature framework that organizes multi-scale descriptors into global, regional, and channel levels to predict dimensional psychopathology. Using the HBN cohort, it shows that tree-based models and granularity-balanced feature selection yield modest improvements, suggesting multi-scale EEG features contain weak but detectable signals for pediatric mental health.