ZUNA1.1可灵活重建任意长度和位置的脑电数据，性能超越传统方法。
ZUNA1.1: A more flexible EEG foundation model for Denoising and Super-resolution
- 基于扩散自编码器，支持任意通道数与时间区间重建
- 在30秒长信号上表现不逊于前代模型，显著优于球面样条插值
- 开源免费，适合脑电去噪与超分辨率研究者使用
我们提出ZUNA1.1，一个380M参数的扩散自编码器，用于灵活的脑电信号重建。该模型可处理长达30秒的变长序列，支持任意数量的脑电通道及任意头皮位置，并能重建通道内的任意时间区间，而不仅限于整通道。实验表明，ZUNA1.1在性能上至少与先前的ZUNA模型相当，且灵活性显著提升，可应对多种重建任务。同时，其在脑电去噪与重建任务中仍显著优于主流方法，如MNE工具包中广泛使用的球面样条插值（spherical spline interpolation）。ZUNA1.1已以宽松的Apache 2.0许可证开源。
原文摘要 · Abstract (English)
We introduce ZUNA1.1, a 380M-parameter diffusion autoencoder for flexible EEG signal reconstruction. ZUNA1.1 is capable of reconstructing variable length sequences of up to 30s, with an arbitrary number of EEG channels at arbitrary scalp locations, and can reconstruct arbitrary temporal intervals within channels in addition to reconstructing entire channels. We demonstrate that ZUNA1.1 performs at least on par with our earlier ZUNA1 model, while being far more flexible and capable of handling a wide range of reconstruction tasks. ZUNA1.1 continues to substantially outperform standard EEG denoising and reconstruction methods such as spherical spline interpolation, which is ubiquitously deployed in the MNE package. The ZUNA1.1 model is released open source under the permissive Apache 2.0 license.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。