The study evaluates how low‑precision compression affects adversarial robustness in EEG decoders used for brain‑computer interfaces. Using BCI Competition IV‑2a data, the authors compare 32‑bit floating‑point models (EEGNet and ShallowConvNet) with models pruned to 50 % and quantized to INT8 via post‑training quantization (PTQ) or quantization‑aware training (QAT). Results show that accuracy‑preserving compression does not improve direct robustness—PGD attack accuracy remains 22–24 % across all variants—yet pruning reduces bidirectional transfer efficiency more than PTQ, indicating that robustness, transferability, and deployment efficiency are distinct properties of compressed EEG decoders.
By Saim Rehman, Muhammad Shafique
ThinkNet is a validation‑controlled framework that uses train‑only normalization, evolutionary search guided by validation, and validation‑gated inference to build compact motor‑imagery EEG decoders that generalize to unseen subjects. In a fixed benchmark on the BCI Competition IV‑2a dataset, the best single model achieved 44.35 % accuracy with only 4.9 k parameters, while a compact ensemble of six families reached 43.31 % accuracy. Across a broader search, models with ≤25 k parameters outperformed larger ones after retraining, but validation reliability remains a bottleneck under subject shift.
By Abdul Basit, Saim Rehman, Muhammad Shafique
arXiv:2608. 07033v1 Announce Type: new Abstract: This work investigates whether Electroencephalograph (EEG) foundation models (EFMs) can be made faster and locally deployable without sacrificing accuracy.
By Lingwei Li, Yirong Kan, Peng Chen, Xu Cao, Zheng Chen, Yasuhiko Nakashima
TriRouter is a new electrode‑space routing framework designed to improve motor‑imagery brain‑computer interface decoding from multichannel EEG when electrode montages are sparse or corrupted. It incorporates discriminative‑preservation, mask‑robustness, and latent‑compensation paths, and uses a validation‑calibrated router that allocates path contributions based on observable channel availability. In the BNCI2014‑001 dataset, TriRouter outperforms standard public backbones and a dynamic‑spatial‑filtering corrupted‑EEG baseline, achieving accuracy gains of 3.68, 4.02, and 6.46 percentage points at 25%, 50%, and 75% missing settings, and remains effective under structured regional missingness and low‑channel evaluations.
By Ying Zhou, Yaoxin, Mengting Liu
arXiv:2609.23924v1 Announce Type: new
Abstract: Pretrained EEG foundation models are increasingly proposed as general-purpose encoders for brain-computer interfaces, yet recent benchmarks disagree ab...
By Kevin Zhou, Sparsh Roy
arXiv:2606. 08594v1 Announce Type: new Abstract: Deep learning EEG denoising architectures have scaled from tens of thousands to tens of millions of parameters, yet no prior study has isolated model capacity as the experimental variable or tested whether reconstruction metrics predict downstream neural-signal utility.
By Jasmeet Singh Bindra, Siddharth Panwar, Shubhajit Roy Chowdhury