Meta-learning In-Context Enables Training-Free Cross Subject Brain Decoding
2026-04-09 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a new method to read visual information from brain scans (fMRI) that works well across different people without needing extra training for each person. Their approach quickly learns how a new person's brain represents images by using a small number of example brain responses. This method combines information from multiple brain areas to make better guesses about what a person is seeing, and it works even if the brain scans come from different machines or people see different images. Their work moves toward creating general brain-reading models that don’t require precise matching of brain anatomy or shared images between subjects.
fMRIvisual decodingneural representationscross-subject generalizationmeta-learningin-context learningfunctional inversionbrain encoding modelnon-invasive brain decodingcomputer vision
Authors
Mu Nan, Muquan Yu, Weijian Mai, Jacob S. Prince, Hossein Adeli, Rui Zhang, Jiahang Cao, Benjamin Becker, John A. Pyles, Margaret M. Henderson, Chunfeng Song, Nikolaus Kriegeskorte, Michael J. Tarr, Xiaoqing Hu, Andrew F. Luo
Abstract
Visual decoding from brain signals is a key challenge at the intersection of computer vision and neuroscience, requiring methods that bridge neural representations and computational models of vision. A field-wide goal is to achieve generalizable, cross-subject models. A major obstacle towards this goal is the substantial variability in neural representations across individuals, which has so far required training bespoke models or fine-tuning separately for each subject. To address this challenge, we introduce a meta-optimized approach for semantic visual decoding from fMRI that generalizes to novel subjects without any fine-tuning. By simply conditioning on a small set of image-brain activation examples from the new individual, our model rapidly infers their unique neural encoding patterns to facilitate robust and efficient visual decoding. Our approach is explicitly optimized for in-context learning of the new subject's encoding model and performs decoding by hierarchical inference, inverting the encoder. First, for multiple brain regions, we estimate the per-voxel visual response encoder parameters by constructing a context over multiple stimuli and responses. Second, we construct a context consisting of encoder parameters and response values over multiple voxels to perform aggregated functional inversion. We demonstrate strong cross-subject and cross-scanner generalization across diverse visual backbones without retraining or fine-tuning. Moreover, our approach requires neither anatomical alignment nor stimulus overlap. This work is a critical step towards a generalizable foundation model for non-invasive brain decoding.