arXiv:2406. 07775v2 Announce Type: replace Abstract: Multimode optical fibres are hair-thin strands of glass that efficiently transport light.
arXiv:2603. 18846v3 Announce Type: replace-cross Abstract: Foundation models are used to extract transferable representations from large amounts of unlabeled data, typically via self-supervised learning (SSL).
The study investigates how aligning the representational geometry of artificial neural networks can improve bidirectional predictivity with biological neural responses. By applying spectral regularization during training of self‑supervised contrastive vision models, the authors increased reverse predictivity by 55% while only modestly reducing forward predictivity. The adjustments also lowered effective dimensionality and reorganized the shared representational subspace, making forward and reverse predictivity more symmetric at intermediate spectral exponents.
arXiv:2607. 15693v1 Announce Type: cross Abstract: We describe a model of perceptual inference in primary visual cortex (V1) equivalent to a minimal diffusion model whose function can be readily understood from its parameters.
arXiv:2606. 03262v1 Announce Type: new Abstract: Neural operators learn mappings between infinite-dimensional function spaces and provide a data-driven surrogate modeling paradigm for parametric partial differential equations (PDEs).
arXiv:2607. 25576v1 Announce Type: new Abstract: Photoacoustic tomography (PAT) combines the optical absorption contrast of biological tissue with the spatial resolution of ultrasound, yet recovering the initial pressure distribution from sparse-view sensor measurements remains an ill-posed inverse problem.
The study investigates how the geometry of representations in artificial neural networks can be steered to improve bidirectional alignment with biological neural responses. By applying spectral regularization during training of self‑supervised contrastive vision models, the authors increased reverse predictivity by 55% while only modestly reducing forward predictivity. The changes also lowered effective dimensionality and reorganized the shared subspace, making forward and reverse predictivity more symmetric at certain spectral exponents.
arXiv:2511. 17126v4 Announce Type: replace-cross Abstract: Emerging deep-learning-based lens library pre-training (LensLib-PT) pipeline offers a new avenue for blind lens aberration correction by training a universal neural network, demonstrating strong capability in handling diverse unknown optical degradations.
arXiv:2607. 16295v1 Announce Type: cross Abstract: Mechanistic interpretability has made significant strides in understanding neural network representations, with sparse dictionary learning (SDL) methods, most prominently sparse autoencoders, as a central paradigm.
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:2608. 20212v1 Announce Type: new Abstract: High-fidelity removal of eyeglasses from video is a major challenge in facial attribute editing, as the underlying facial geometry is often obscured by complex refractive distortions and view-dependent specular reflections.
The paper presents a method for training single‑step neural surrogates that can handle wave‑scattering inverse problems with tens of thousands of controllable variables. By dynamically generating training examples through gradient ascent and using a replay dataset with normalization, the authors achieve a surrogate that accurately models two‑dimensional wave scattering for up to 41,772 variables and can generalize to over 3 million variables without retraining. The surrogate demonstrates comparable or better performance than traditional FDTD simulations for large‑scale forward simulations and inverse design of photonic devices, achieving speedups up to 26.5×.
arXiv:2606. 14757v1 Announce Type: cross Abstract: Though Vision Transformers (ViTs) have become the dominant backbone in many computer vision tasks, due to permutation equivariance, their attention mechanism lacks explicit spatial inductive biases.