aiqbench analysis
TechPulse take: MindHier introduces a hierarchical autoregressive framework for fMRI-to-image reconstruction that replaces static diffusion guidance with scale-aware neural embeddings. This approach aligns better with cognitive processes by synthesizing global semantics before refining local details, resulting in superior semantic fidelity and a 4.67x speedup over diffusion baselines. The method offers a more deterministic and efficient alternative for bridging neuroscience and computer vision.
Moving Beyond Diffusion: Hierarchy-to-Hierarchy Autoregression for fMRI-to-Image Reconstruction
arXiv:2510.22335v2 Announce Type: replace-cross Abstract: Reconstructing visual stimuli from fMRI signals is a central challenge bridging machine learning and neuroscience. Recent diffusion-based methods typically map fMRI activity to a single neural embedding, using it as static guidance throughout the entire generation process. However, this fixed guidance collapses hierarchical neural information and is misaligned with the stage-dependent demands of image reconstruction. In response, we propose MindHier, a coarse-to-fine fMRI-to-image reconstruction framework built on scale-wise autoregressive modeling. MindHier introduces three components: a Hierarchical fMRI Encoder to extract multi-level neural embeddings, a Hierarchy-to-Hierarchy Alignment scheme to enforce layer-wise correspondence with CLIP features, and a Scale-Aware Coarse-to-Fine Neural Guidance strategy to inject these embeddings into autoregression at matching scales. These designs make MindHier an efficient and cognitively aligned alternative to diffusion-based methods by enabling a hierarchical reconstruction process that synthesizes global semantics before refining local details, akin to human visual perception. Extensive experiments on the NSD dataset show that MindHier achieves superior semantic fidelity, 4.67$\times$ faster inference, and more deterministic results than the diffusion-based baselines.
Abstract:Reconstructing visual stimuli from fMRI signals is a central challenge bridging machine learning and neuroscience. Recent diffusion-based methods typically map fMRI activity to a single neural embedding, using it as static guidance throughout the entire generation process. However, this fixed guidance collapses hierarchical neural information and is misaligned with the stage-dependent demands of image reconstruction. In response, we propose MindHier, a coarse-to-fine fMRI-to-image reconstruction framework built on scale-wise autoregressive modeling. MindHier introduces three components: a Hierarchical fMRI Encoder to extract multi-level neural embeddings, a Hierarchy-to-Hierarchy Alignment scheme to enforce layer-wise correspondence with CLIP features, and a Scale-Aware Coarse-to-Fine Neural Guidance strategy to inject these embeddings into autoregression at matching scales. These designs make MindHier an efficient and cognitively aligned alternative to diffusion-based methods by enabling a hierarchical reconstruction process that synthesizes global semantics before refining local details, akin to human visual perception. Extensive experiments on the NSD dataset show that MindHier achieves superior semantic fidelity, 4.67$\times$ faster inference, and more deterministic results than the diffusion-based baselines.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2510.22335 [cs.CV]
(or arXiv:2510.22335v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2510.22335 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 264 words
Related coverage
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
Import AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop
Import AI · Sep 28, 2026
Humanity’s Last Invention — Richard Socher of Recursive
Latent Space · Sep 14, 2026
Introducing Cosmos 3 Edge
Hugging Face · Jul 20, 2026