![[2511.02565] A Cognitive Process-Inspired Architecture for Subject-Agnostic Brain Visual Decoding](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2511.02565] A Cognitive Process-Inspired Architecture for Subject-Agnostic Brain Visual Decoding
Summary
The paper presents VCFlow, a novel architecture for subject-agnostic brain visual decoding, enhancing the reconstruction of visual experiences from fMRI data without requiring subject-specific training.
Why It Matters
This research addresses the challenges of cross-subject generalization in brain decoding, which is crucial for clinical applications. By improving the efficiency and scalability of visual reconstruction, it opens new avenues for understanding brain activity and developing cognitive technologies.
Key Takeaways
- VCFlow models the human visual system's architecture for improved decoding.
- It utilizes a feature-level contrastive learning strategy for better subject-invariant representations.
- The framework significantly reduces computation time while maintaining accuracy.
Computer Science > Computer Vision and Pattern Recognition arXiv:2511.02565 (cs) [Submitted on 4 Nov 2025 (v1), last revised 24 Feb 2026 (this version, v2)] Title:A Cognitive Process-Inspired Architecture for Subject-Agnostic Brain Visual Decoding Authors:Jingyu Lu, Haonan Wang, Qixiang Zhang, Xiaomeng Li View a PDF of the paper titled A Cognitive Process-Inspired Architecture for Subject-Agnostic Brain Visual Decoding, by Jingyu Lu and 3 other authors View PDF HTML (experimental) Abstract:Subject-agnostic brain decoding, which aims to reconstruct continuous visual experiences from fMRI without subject-specific training, holds great potential for clinical applications. However, this direction remains underexplored due to challenges in cross-subject generalization and the complex nature of brain signals. In this work, we propose Visual Cortex Flow Architecture (VCFlow), a novel hierarchical decoding framework that explicitly models the ventral-dorsal architecture of the human visual system to learn multi-dimensional representations. By disentangling and leveraging features from early visual cortex, ventral, and dorsal streams, VCFlow captures diverse and complementary cognitive information essential for visual reconstruction. Furthermore, we introduce a feature-level contrastive learning strategy to enhance the extraction of subject-invariant semantic representations, thereby enhancing subject-agnostic applicability to previously unseen subjects. Unlike conventional pipelin...
