[论文] Beyond Trial Averaging: Anchoring Neural and Visual Representations fo...

论文概要

研究领域: ML
作者: Zhenyao Cui, Siyuan Kan, Dingkun Liu, Dongrui Wu
发布时间: 2026-08-19
arXiv: 2608.19128

中文摘要

从大脑信号解码视觉信息探测神经表示，并实现神经康复和梦境解码。最近的脑到图像检索方法取得了有前景的性能，通常通过平均每个图像的许多（多达80个）神经试验，需要重复刺激呈现，这增加了延迟、成本和用户负担。当只有一或少量重复可用时，检索准确率急剧下降。这种下降通常归因于查询噪声，因为平均抑制噪声并增加信号稳定性。然而，我们发现一种非传递性对齐模式：低重复查询信号和图像表示各自与高重复中心对齐，但彼此之间不直接对齐。这种模式表明查询噪声只是问题的一部分，gallery放置也影响检索。因此，我们提出基于神经锚的检索（NEAR）框架，将高重复中心视为锚并从两侧接近它：去噪器将噪声查询拉向真实锚，一个小网络预测每个候选者的伪锚并将其图像拉向它。在跨越EEG、MEG和fMRI的四个数据集上，NEAR在少重复制度中一致地改进了检索。在THINGS-EEG2上，当平均1和4次重复时，它分别将200路Top-1准确率提高了5.7和9.3个百分点。通过锚定神经和视觉表示，NEAR减少对重复采集的依赖，并使神经检索更接近实际部署。

原文摘要

Decoding visual information from brain signals probes neural representations and enables neuro-rehabilitation and dream decoding. Recent brain-to-image retrieval approaches have achieved promising performance, typically by averaging many (up to 80) neural trials per image, requiring repeated stimulus presentation that increases latency, cost, and user burden. When only one or a few repetitions are available, the retrieval accuracy drops sharply. This drop is commonly attributed to query noise because averaging suppresses noise and increases signal stability. However, we find a non-transitive alignment pattern: the low-repetition query signal and the image representation each align with the high-repetition center, but not directly with each other. This pattern shows that query noise is only...

自动采集于 2026-08-21

#论文 #arXiv #ML #小凯