MindAdapter: Few-Shot Parameter-Efficient Residual Calibration of Cross-Subject Brain-to-Visual Decoding Models

2026-05-23

AI Summary

This work addresses the systematic alignment bias in cross-subject brain-to-vision decoding caused by inter-individual functional variability. To tackle this challenge, the authors propose MindAdapter, a framework that introduces lightweight nonlinear residual adapters while keeping both the pretrained brain functional alignment backbone and the vision-language decoder frozen. This design enables a decoupled linear–residual cascaded alignment paradigm, facilitating fine-grained spatial and semantic calibration. Innovatively, the method incorporates topologically anchored dual-stream manifold constraints, which, under minimal shared stimulus conditions, jointly preserve global representational stability and inject subject-specific corrections. Experiments on the Natural Scenes Dataset (NSD) demonstrate that MindAdapter significantly improves cross-subject visual reconstruction and retrieval performance, achieving efficient and personalized brain decoding.

Abstract

Cross-subject brain-to-visual decoding remains a core challenge in brain-computer interfaces due to severe inter-individual variability that induces systematic subject-specific functional misalignment. To address this issue, we propose MindAdapter, a parameter-efficient few-shot calibration framework for pretrained brain-to-visual decoding models. MindAdapter adopts a decoupled linear-residual cascade alignment paradigm by freezing a pretrained explicit brain functional alignment backbone (coarse) and introducing a lightweight nonlinear residual adapter (fine), thereby disentangling global cross-subject correspondence from subject-specific residual corrections for fine-grained spatial and semantic calibration. To further preserve global representational stability, we design a topology-anchored dual-stream manifold constraint, where a small set of shared stimuli serves as topological pins with voxel-level paired supervision, while a semantic stream enforces consistency through a frozen vision-language decoder on unpaired brain data. Together, MindAdapter efficiently injects subject-specific corrections while maintaining the global representational geometry learned during pretraining. Experiments on the Natural Scenes Dataset (NSD) demonstrate that MindAdapter substantially improves cross-subject visual reconstruction and retrieval accuracy using only a few shared stimuli, offering a practical and data-efficient solution for personalized brain-to-visual decoding.

Problem

• cross-subject
• brain-to-visual decoding
• inter-individual variability
• functional misalignment
• few-shot calibration

Innovation

• parameter-efficient adaptation
• few-shot calibration
• cross-subject alignment
• residual adapter
• manifold constraint