About
Research & Development
A New Foundation Model for Particle Physics Data
TLDR: FM4NPP is a new foundation model for nuclear and particle physics that uses a novel self-supervised training method for sparse detector data. Trained...
Research & Development
Uncovering Hidden Communities in Complex Networks with a New Graph Neural Network
TLDR: A new Graph Neural Network called uAGNN is introduced for unsupervised community detection in graphs. It excels in "heterophilic" graphs, where similar nodes...
Research & Development
New AI Model Enhances Understanding of High-Order Relationships in Data
TLDR: A new research paper introduces the Implicit Hypergraph Neural Network (IHNN), the first implicit model for hypergraphs. IHNN addresses the limitation of existing...
Research & Development
Assessing AI-Generated Software: A New Approach to Interactive Evaluation
TLDR: RealDevWorld is a novel framework for evaluating AI-generated software, addressing the limitations of current methods that struggle with interactive applications. It features RealDevBench,...
Research & Development
A New Dataset for Advancing AI in Solar Physics and Space Weather Forecasting
TLDR: SuryaBench is a new, high-resolution, machine learning-ready dataset derived from NASA's Solar Dynamics Observatory (SDO). It provides preprocessed solar imagery (AIA and HMI)...
Research & Development
Federated Learning Enhances Worker Action Recognition in Smart Manufacturing
TLDR: This research paper introduces a federated learning (FL) framework for pose-based human activity recognition (HAR) in smart manufacturing. Using a custom dataset of...