Towards Interpretable Visual Decoding with Attention to Brain Representations

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Feng, Pinyuan, Adeli, Hossein, Guo, Wenxuan, Cheng, Fan, Hwang, Ethan, Kriegeskorte, Nikolaus
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917301136130048
author Feng, Pinyuan
Adeli, Hossein
Guo, Wenxuan
Cheng, Fan
Hwang, Ethan
Kriegeskorte, Nikolaus
author_facet Feng, Pinyuan
Adeli, Hossein
Guo, Wenxuan
Cheng, Fan
Hwang, Ethan
Kriegeskorte, Nikolaus
contents Recent work has demonstrated that complex visual stimuli can be decoded from human brain activity using deep generative models, offering new ways to probe how the brain represents real-world scenes. However, many existing approaches first map brain signals into intermediate image or text feature spaces before guiding the generative process, which obscures the contributions of different brain areas to the final reconstruction output. In this work, we propose NeuroAdapter, a visual decoding framework that directly conditions a latent diffusion model on brain representations, bypassing the need for intermediate feature spaces. Our method demonstrates competitive visual reconstruction quality on public fMRI datasets compared to prior work, while providing greater transparency into how brain signals drive visual reconstruction. To this end, we introduce an Image-Brain BI-directional interpretability framework (IBBI) that analyzes cross-attention patterns across diffusion denoising steps to reveal how different cortical areas influence the unfolding generative trajectory. Our work highlights the potential of end-to-end brain-to-image reconstruction and establishes a path for interpretable neural decoding.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23566
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Interpretable Visual Decoding with Attention to Brain Representations
Feng, Pinyuan
Adeli, Hossein
Guo, Wenxuan
Cheng, Fan
Hwang, Ethan
Kriegeskorte, Nikolaus
Computer Vision and Pattern Recognition
68T07
I.2.0; I.4.9
Recent work has demonstrated that complex visual stimuli can be decoded from human brain activity using deep generative models, offering new ways to probe how the brain represents real-world scenes. However, many existing approaches first map brain signals into intermediate image or text feature spaces before guiding the generative process, which obscures the contributions of different brain areas to the final reconstruction output. In this work, we propose NeuroAdapter, a visual decoding framework that directly conditions a latent diffusion model on brain representations, bypassing the need for intermediate feature spaces. Our method demonstrates competitive visual reconstruction quality on public fMRI datasets compared to prior work, while providing greater transparency into how brain signals drive visual reconstruction. To this end, we introduce an Image-Brain BI-directional interpretability framework (IBBI) that analyzes cross-attention patterns across diffusion denoising steps to reveal how different cortical areas influence the unfolding generative trajectory. Our work highlights the potential of end-to-end brain-to-image reconstruction and establishes a path for interpretable neural decoding.
title Towards Interpretable Visual Decoding with Attention to Brain Representations
topic Computer Vision and Pattern Recognition
68T07
I.2.0; I.4.9
url https://arxiv.org/abs/2509.23566