Neural Networks

A developer outlines a strategy for on-device training of lightweight neural networks in iOS apps, enabling personalized predictions while preserving user privacy. The approach uses a compact MLP trained locally on user …

Larkos 0.3 introduces a GAT-based neuron graph reasoner, a temporal attention encoder, and a refactored fusion head. The update replaces mean-pooling with learned-query attention and increases memory capacity, while the …

A new coding implementation demonstrates how to infer urban functions using spatial graph neural networks. The pipeline collects POI and street network data from OpenStreetMap, constructs proximity graphs, and trains a G…

A researcher applied activation patching to a DNA sequencing basecaller, a deep learning model, to understand its internal mechanisms, finding MLP dominance in early and late layers and concentrated self-attention activi…

A developer explains the sparse Mixture of Experts (MoE) architecture used in models like Mixtral, DeepSeek-MoE, and Grok-1, detailing how the router selects which experts to activate per token and why load-balancing is …

A Nature preprint published June 13, 2026, introduces CURE, a graph unlearning framework that uses contrastive representation editing to remove the influence of target nodes, edges, or features from a trained graph neura…

TONE3000 released Neural Amp Modeler Architecture 2 (A2) on June 2, an open-source update developed with creator Steve Atkinson that the company claims outperforms competing modelling systems in blind listening tests whi…

Researchers at Fourthline, a KYC identity verification company, observed that AI models trained with supervised versus contrastive loss functions produce different prediction scores when the input video codec is changed.…

Arm demonstrated its neural technologies for mobile phone gaming with a ray-traced demo, showcasing an alternative chip for handheld gaming PCs beyond AMD or Intel. The presentation highlighted the potential for Arm-base…

MONAI, an open-source medical imaging framework, has released a tutorial demonstrating an end-to-end 3D spleen segmentation pipeline using a UNet model on CT volumes from the Medical Segmentation Decathlon Task09 dataset…

Two ESP32-S3 microcontrollers running a Llama-architecture language model have achieved the first multi-chip pipelined LLM inference on ESP32-class hardware, splitting layers across two boards connected by three jumper w…

Researchers introduced SupraSNN, a hardware-software co-design that treats synaptic events as parallelizable micro-operations and physically decouples synaptic and neuronal computation, achieving synapse-level parallelis…

Shaivi Malik published an arXiv paper on 11 June 2026 that frames neural model editing as a reinforcement learning problem, using reward feedback to train agents that modify pretrained networks. The paper introduces two …

Researchers have developed Context-Centric Feature Fusion (CCFF), a framework that uses two attention-based modules to improve detection of co-occurring objects in autonomous driving. The system achieved a Category-level…

Researchers have introduced Rotary Value Embeddings (RoVE), a parameter-free modification to Rotary Position Embeddings (RoPE) that makes value pathways position-sensitive by rotating values simultaneously with keys. Thi…

Researchers have developed a Density-Equalizing Convolutional Neural Network (DECNN) that uses learned density functions to dynamically redistribute computational attention toward informative regions in images and surfac…

Researchers have developed YOLO-AMC, an improved YOLO architecture integrating attention mechanisms for building crack detection, achieving mAP@0.5 of 0.9917 on test datasets. The model outperforms baseline YOLOv11n and …

Researchers have developed CD-RCM, the first novel-view synthesis approach specifically for reflectance confocal microscopy (RCM), enabling continuous-depth visualization of human skin from sparse, anisotropic z-stacks. …

Researchers at an undisclosed institution reverse-engineered Apple's Metal 4.1 tensor compute path on the M4 Max GPU, revealing that the fp8 matmul2d operation is emulated rather than hardware-accelerated. The study, pub…

Researchers at the University of Oxford found that observable patterns in latent reasoning models, such as BFS-like frontiers and decodable arithmetic, also appear in control models lacking the proposed recurrence or cur…