EVENT DOSSIER
On September 7, 2026, arXiv cs.AI published the paper “ProCA: Progressive Contrastive Alignment for Robust EEG Visual Decoding”. This study introduced the ProCA framework, which optimizes EEG signal processing through a progressive contrastive alignment strategy. The method aims to address the issues of high noise interference and unstable feature extraction in traditional EEG decoding, thereby improving the robustness and accuracy of visual decoding.
arXiv:2609.05094v1 提出 ProCA（Progressive Contrastive Alignment），一种针对 EEG 视觉解码的自适应神经语义对齐框架。现有方法依赖固定视觉或文本锚点，导致其语义关系随试次、受试者及学习阶段变化而错位，引发鲁棒性下降。ProCA 通过从冻结的视觉语言先验逐步细化至 EEG 感知语义关系，并结合基于通道与时序重要性的结构一致插值约束特征混合。在受试者依赖、独立、严格跨受试者迁移及持续适应等设置下，该方法分别实现平均 Top-1/Top-5 相对增益为 7.4%/3.9%、10.0%/4.6%、28.1%/17.8% 和 16.8%/11.6%。