---
title: "超越试验平均：锚定神经与视觉表征实现少重复脑到图像检索"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.19128v1"
permalink: "https://feed.astralor.com/items/itm_4095fe063a6f95aeddd2621c"
published: "2026-08-19T17:23:30.000Z"
score: 78
channel: "ai"
tags: ["NEAR","THINGS-EEG2","少样本学习","神经解码","脑到图像检索","脑机接口"]
---

# 超越试验平均：锚定神经与视觉表征实现少重复脑到图像检索

## AI 点评

> 这篇论文真正有价值的地方在于它把'少量重复导致检索崩坏'这个被当成噪声问题处理的现象重新拆解，发现了一个更微妙的表征对齐结构。对做脑机接口或神经解码的人来说，NEAR可能意味着未来实验不需要让被试盯着图片看几十遍，直接就能从单次信号里找回图像。虽然预印本还没经过正式评审，但方法本身跨了三类神经成像数据，说服力很强。

## 速览

脑到图像检索常需对每张图片重复采集多达80次神经信号并取平均，以抑制噪声；当重复次数降至1或4次时，检索精度大幅下降。本文发现低重复查询信号与图像表征各自对齐高重复中心，但彼此并不直接对齐，说明问题不仅源于查询噪声，也与图库分布有关。为此提出NEAR框架，将高重复中心视为锚点，从两侧逼近：去噪器将带噪查询拉向真实锚点，小网络从图像预测伪锚点并拉近图像。在EEG、MEG和fMRI四个数据集上，NEAR一致提升少重复场景下的检索性能，在THINGS-EEG2上200路Top-1准确率分别提升5.7和9.3个百分点（平均1次和4次）。该工作降低了对重复采集的依赖，推动神经检索走向实际应用。

## 关键点

- 发现低重复查询与图像表征各自对齐高重复中心，但彼此不直接对齐的非传递模式，揭示图库放置同样影响检索。
- 提出NEAR框架，利用高重复中心作锚点，从查询侧去噪和从图像侧预测伪锚点双向逼近。
- 在EEG、MEG、fMRI四个数据集上验证，少重复场景下检索性能一致提升。
- 在THINGS-EEG2上，1次和4次平均时200路Top-1准确率分别提升5.7和9.3个百分点。

## 全文

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 part of the problem and that gallery placement also affects retrieval. We therefore propose a neural-anchor-based retrieval (NEAR) framework that treats the high-repetition center as an anchor and approaches it from both sides: a denoiser pulls the noisy query toward the true anchor, and a small network predicts each candidate's pseudo anchor from its image and pulls the image toward it. Across four datasets spanning EEG, MEG and fMRI, NEAR consistently improved retrieval in the few-repetition regime. On THINGS-EEG2, it improved 200-way Top-1 accuracy by 5.7 and 9.3 percentage points respectively, when averaging one and four repetitions. By anchoring neural and visual representations, NEAR reduces reliance on repeated acquisition and brings neural retrieval closer to real-world deployment.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_4095fe063a6f95aeddd2621c