aiqbench analysis
This research demonstrates a method for steering the spectral geometry of neural representations in self-supervised vision models to achieve more symmetric alignment between biological and artificial systems. By balancing forward and reverse predictivity, the study offers new insights into bridging the gap between AI architectures and human brain function without sacrificing model performance.
Bidirectional representational alignment between biological and artificial neural networks
arXiv:2608.18244v1 Announce Type: cross Abstract: Recent work has shown that representational alignment between biological and artificial neural networks is asymmetric: model representations predict neural responses much better than neural responses predict model representations. This asymmetry raises the question of whether representational geometry contributes to bidirectional representational alignment. We hypothesized that steering representational geometry during training can systematically influence bidirectional alignment. To test this hypothesis, we developed a computational framework that integrates spectral regularization with bidirectional predictivity analyses. As an initial demonstration, we evaluated our framework using self-supervised contrastive vision models. Steering the spectral geometry of the learned representations substantially increased reverse predictivity with modest reductions in forward predictivity, yielding a 55% relative improvement in bidirectional predictivity. These improvements were accompanied by reduced effective dimensionality and a reorganization of the shared representational subspace, within which forward and reverse predictivity became approximately symmetric at intermediate spectral exponents. Overall, these findings demonstrate that representational geometry can be systematically steered to modulate bidirectional representational alignment between biological and artificial neural networks.
Abstract:Recent work has shown that representational alignment between biological and artificial neural networks is asymmetric: model representations predict neural responses much better than neural responses predict model representations. This asymmetry raises the question of whether representational geometry contributes to bidirectional representational alignment. We hypothesized that steering representational geometry during training can systematically influence bidirectional alignment. To test this hypothesis, we developed a computational framework that integrates spectral regularization with bidirectional predictivity analyses. As an initial demonstration, we evaluated our framework using self-supervised contrastive vision models. Steering the spectral geometry of the learned representations substantially increased reverse predictivity with modest reductions in forward predictivity, yielding a 55% relative improvement in bidirectional predictivity. These improvements were accompanied by reduced effective dimensionality and a reorganization of the shared representational subspace, within which forward and reverse predictivity became approximately symmetric at intermediate spectral exponents. Overall, these findings demonstrate that representational geometry can be systematically steered to modulate bidirectional representational alignment between biological and artificial neural networks.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2608.18244 [cs.LG]
(or arXiv:2608.18244v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.18244 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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 · 259 words
Related coverage
The J-Space Debate, Agent Swarms, and Pacing Frontier AI - Digital Minds Newsletter #4
LessWrong · Sep 18, 2026
Not all solutions are created equal: An analytical dissociation of functional and representational similarity in deep linear neural networks
arXiv · Oct 1, 2026
Neuroscience Institute researchers use AI to model brain’s image sense.
Quantum Zeitgeist · Aug 30, 2026
Neural networks are more modular than single neurons suggest
arXiv · Oct 12, 2026