Neural Networks news and analysis on Web Pulse: 1659 curated articles tracking the latest Neural Networks developments, tools, and research, updated continuously from vetted sources.
A September 11, 2026 arXiv paper shows that permutation-based model confidentiality in hybrid fully homomorphic encryption (FHE) inference fails, with d+1 admissible queries per linear layer enabling exact recovery of a …
A physiology-guided hybrid framework combining a convolutional neural network branch with a morphology-prior branch estimated blood pressure from photoplethysmography (PPG) with mean absolute errors of 3.77 mmHg for syst…
Researchers proposed a Position Encoding-Based Deformable Spatial Aggregation Module (PEBDSAM) that addresses over-smoothing, over-compression, fixed receptive fields, and topological noise in graph neural networks, acco…
Researchers introduced MACCHIATO, a specialized training algorithm that jointly constructs an explicitly structured ReLU-MLP from partial truth-table observations and an explicit Boolean circuit over signed literals with…
A new arXiv paper, arXiv:2609.13204v1, introduces GradRepair-ODE, a reliability framework that checks, repairs, and rejects gradients in neural ODE training at the optimizer step. Across six synthetic ODE systems, GradRe…
Researchers introduced embedding-based partial information decomposition (ePID), a scalable pipeline that compresses nonfocal symptoms into a low-cardinality discrete embedding to compute a tractable two-source PID, acco…
Researchers introduced SPICE (Simple Polysemantic Feature Interpretation via Clustering-based Explanation), a framework for analyzing polysemanticity in deep vision architectures, detailed in arXiv paper 2609.13198v1. SP…
A paper posted to arXiv (arXiv:2609.13171v1) formulates derivative fidelity as a failure mode of physics-informed neural networks (PINNs), showing that multilayer perceptrons trained only on function values for sin(x) an…
Samsung Electronics and Verizon Communications completed an AI-powered Integrated Sensing and Communication (ISAC) trial at a major international soccer fan event in Texas, Samsung announced Tuesday, calling it one of th…
A technical reference on transformer architecture details that most tokenizers prepend a BOS token, which acts as a fixed landmark or "attention sink," while a space prefix yields the wrong token. The reference notes tha…
Sakana AI researchers Jeffrey Seely and Julian Gould introduced Augmented Lagrangian Predictive Coding (PC-ALM), a layer-local alternative to backpropagation that attaches a Lagrange multiplier to each layer constraint. …
Transolver, UPT, and AB-UPT each replace full point-to-point self-attention with smaller token sets to make global computation affordable on industrial-scale meshes, according to Part 5 of the Geometric Deep Learning ser…
A developer released neural_texture_es2, a self-contained C++ experiment that encodes an RGB image or up to four PBR material textures as a shared low-resolution latent texture plus a tiny MLP decoder, trained entirely w…
Researchers Florian Hahlbohm, Jorge Condor, Linus Franke, Martin Eisemann and Marcus Magnor published an arXiv paper introducing a compact neural appearance model for 3D Gaussian Splatting that reduces the per-primitive …
A developer released a from-scratch Sudoku solver built on a Looped MLP-Mixer architecture, trained on the sapientinc/sudoku-extreme dataset and runnable on an NVIDIA Jetson Nano via JetPack Docker. The model reuses a sh…
Developer ErnestoSOFTWARE released an interactive browser project that simulates a fruit fly's male central nervous system connectome playing chess against a human opponent, rendering live neural activity across the opti…
A blog post explains the Residual-Quantized Variational Autoencoder (RQ-VAE), an architecture that extends the Vector Quantized Variational Autoencoder (VQ-VAE) introduced in arXiv paper 1711.00937 to capture hierarchica…
Researchers introduced Fundamental Dynamical Units (FDUs), signed three-node interaction patterns, as composable primitives for recovering interaction structure from perturbation time-series data in networked dynamical s…
Researchers introduced Diffusion-Conditioned Representation Alignment (DCRA), a training framework that repurposes the forward diffusion process as a structured corruption scheduler for time-series representation learnin…
A multifidelity railway-bogie response-correction method combining a time-delay neural network (TDNN) with physics-informed residual learning achieved a mean coefficient of determination of 0.8197, a mean normalized root…