---
title: "SCORE: Subject Coordinate Recovery for Label-Free Cross-Subject EEG-to-Image Retrieval"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.19134v1"
permalink: "https://feed.astralor.com/items/itm_425dd9dbbbd95323ab8b32c1"
published: "2026-08-19T17:27:29.000Z"
score: 78
channel: "ai"
tags: ["SCORE","Image Retrieval","EEG","Visual Decoding","Zero-shot Adaptation"]
---

# SCORE: Subject Coordinate Recovery for Label-Free Cross-Subject EEG-to-Image Retrieval

## AI commentary

> This work neatly dissects the problem, focusing on cross-subject coordinate shifts rather than stacking complex networks. The label-free coordinate alignment approach yields substantial gains, making it a valuable reference for BCI and cross-user retrieval researchers, and a step toward plug-and-play neural retrieval for product teams.

## At a glance

EEG-to-image retrieval holds great promise for neural communication, but performance drops sharply for new users without labeled calibration. This paper analyzes EEG features and finds that different subjects preserve similar concept relationships but express them along different coordinate directions, leading to the Subject Coordinate Recovery (SCORE) framework. During training, SCORE aligns source-subject EEG with a common image space and simulates unseen-subject recovery via source-only episodes. At deployment, with frozen encoders, it selects reliable EEG-image landmarks through hubness-corrected matching and estimates an orthogonal transformation to recover target coordinates without source data or target labels. On THINGS-EEG2 and Alljoined-1.6M, SCORE outperforms the strongest baselines by significant margins in 200-way retrieval, bringing practical cross-user visual decoding closer to deployment.

## Key points

- Identifies that EEG features across subjects share concept structure but differ in coordinate directions, explaining the cross-subject performance gap.
- Proposes SCORE, which performs zero-shot cross-subject retrieval via coordinate alignment without target labels or encoder updates.
- Achieves 53.23% and 12.01% Top-1 accuracy on THINGS-EEG2 and Alljoined-1.6M, surpassing strongest baselines by 17.45 and 3.08 percentage points.

## Full text

Accurate visual decoding can reveal how the brain represents visual information and recover perceived content from neural signals such as electroencephalography (EEG), with potential for neural communication. However, current EEG-to-image retrieval methods perform far below their within-subject counterparts for new users without labeled calibration, limiting real-world deployment. To understand this gap, we analyze EEG features across subjects and find that different subjects preserve similar relationships among concepts but express them along different coordinate directions. We therefore propose Subject Coordinate Recovery (SCORE), a target label-free framework combining recovery-aware source training with coordinate alignment at deployment. During training, SCORE aligns source subject EEG with a common image space and simulates unseen-subject recovery through source-only episodes. At deployment, with both encoders frozen, SCORE selects reliable EEG-image landmarks through hubness-corrected matching and estimates an orthogonal transformation to recover target EEG coordinates without source data or target labels. In 200-way retrieval on two public benchmarks, SCORE outperforms the unadapted baseline for every target subject and achieves the best overall accuracy. It reaches 53.23%/83.55% and 12.01%/32.16% Top-1/Top-5 on THINGS-EEG2 and Alljoined-1.6M, respectively, surpassing the strongest baselines by 17.45/15.70 and 3.08/4.62 percentage points. Without target labels or encoder updates, SCORE brings brain-based visual decoding closer to robust, practical, low-latency deployment across users.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_425dd9dbbbd95323ab8b32c1