arXiv:2607. 00025v1 Announce Type: cross Abstract: While deep learning models achieve state-of-the-art performance in complex tasks, they remain brittle when faced with new environments or sensory deprivation.
arXiv:2610.08418v1 Announce Type: new Abstract: Biological connectomes encode structured solutions to visual computation that may provide reusable inductive biases for artificial vision. We develop C...
arXiv:2606. 14975v1 Announce Type: cross Abstract: How the wiring and functional organization of cortex shape recurrent computation remains a central question in both neuroscience and machine learning.
Can a biological wiring diagram serve as a useful computational substrate beyond the behaviors for which it evolved? We use the publicly released MaleCNS v1.0 connectome, reconstructed from a single a...
The study investigates whether a biological wiring diagram can serve as a computational substrate for tasks beyond its evolutionary purpose. Using the MaleCNS v1.0 Drosophila connectome as a fixed recurrent network, the authors train models to perform bounded addition and a grounded relational language task, achieving high accuracy (92.77% for addition and 61.59% for language) that surpasses degree-preserving rewired controls. The results demonstrate that the higher-order structure of the MaleCNS provides a reusable inductive bias for these cognitive tasks.
arXiv:2603. 01568v2 Announce Type: replace Abstract: Efficient coding theory predicts that biological perceptual systems compress sensory input optimally under resource constraints, with the systematic structure of errors reflecting the geometry of that compression.
arXiv:2609.24565v2 Announce Type: replace Abstract: A connectome-constrained model of the fly visual system, optimized for motion and then frozen, can be driven over architectural drawings by prescri...
Mechanistic interpretability of vision transformers seeks to decompose model computation into human-readable units, but learned representations entangle many concepts in each neuron. Feature superposi...
arXiv:2609.24379v1 Announce Type: cross Abstract: Mechanistic interpretability of vision transformers seeks to decompose model computation into human-readable units, but learned representations entan...
arXiv:2609.24565v1 Announce Type: new Abstract: Architectural drawings encode material classes through repeated hatch patterns. We test whether a connectome-constrained fly visual network, pretrained...
arXiv:2602. 17997v3 Announce Type: replace Abstract: Animals perform coordinated whole-body movements under the control of neural systems shaped by brain-wide connectivity.
The paper introduces a deep learning pipeline that rapidly and accurately registers 3D high‑resolution Drosophila larval brain volumes to a shared anatomical reference. Unlike traditional methods that require per‑case optimization and minutes per brain, the trained network performs a single forward pass, handling volumes with many more voxels and maintaining high accuracy even as image quality declines. The authors benchmarked their approach against eleven classical and seven learned baselines, achieving a 23‑percentage‑point improvement in landmark‑based mutual information and registering brains one to two orders of magnitude faster.