An optimized implementation of the Kimi Linear architecture - a hybrid linear attention mechanism outperforming traditional full attention.
-
Updated
Jun 13, 2026 - Python
An optimized implementation of the Kimi Linear architecture - a hybrid linear attention mechanism outperforming traditional full attention.
A comprehensive toolkit for developing and benchmarking compression algorithms specifically designed for neural data streams in brain-computer interfaces (BCIs). This project provides efficient, real-time compression solutions that preserve the critical characteristics
(NeurIPS 2025), Compact CNNs for EEG decoding: response time prediction and behavioral assessment using competition starter kit infrastructure with custom normalization and training strategies using the Healthy Brain Network (HBN) EEG dataset
This project demonstrates a comprehensive machine learning pipeline with examples of supervised, unsupervised, and semi-supervised learning approaches. It serves as a template and learning resource for ML practitioners.
This project implements a deep learning pipeline for tumor detection and segmentation in medical images (MRI/CT) using the MONAI framework and PyTorch.
A comprehensive industrial automation platform demonstrating integration between machine vision systems, industrial robots, PLCs, and quality control systems for modern manufacturing applications.
An AI-powered system for analyzing James Webb Space Telescope images to identify artificial structures, Dyson spheres, and objects that don't follow standard gravitational rules - potential indicators of intelligent extraterrestrial life.
This repository contains experimental quantum computing algorithms and simulations for cutting-edge research applications including medical genomics, cosmology, and quantum machine learning.
A comprehensive Python-based machine learning platform for real-time seismic event detection, analysis, and classification. This system integrates with authoritative seismic data sources (USGS and IRIS) to provide intelligent earthquake monitoring and analysis capabilities.
A project to build GPU acceleration for LLaMA models on local computers and AWS, leveraging GPU resources for efficient inference and training.
QuantumForge is an open-source framework that revolutionizes quantum chemistry calculations by combining the power of GPU acceleration, deep learning, and density functional theory. Built for researchers who demand both accuracy and performance.
A comprehensive Python-based simulation environment for First Lego League competitions, featuring realistic robot physics, interactive game maps, and mission scenarios.
A specialized compression and interface layer that enables Apple's BCI HID technology to work more efficiently with existing BCI compression algorithms, focusing on low latency and high signal quality.
Lightweight, extensible Brownian dynamics toolkit for nanoparticles and proto-nanorobotics NanoSimLab provides accessible tools for simulating and analyzing nanoparticle systems using Brownian dynamics, with a focus on nanorobotics research and development. The toolkit runs out-of-the-box with NumPy/SciPy and offers seamless integration.
AdaAttn is a GPU-native attention mechanism that dynamically adapts both numerical precision and matrix rank at runtime, reducing memory bandwidth and computational overhead in large language models without sacrificing model quality. By aligning linear algebra operations with modern GPU hardware characteristics.
Python project to experiment with bounded spacecraft motion near the binary asteroid system Moshup-Squannit (1999 KW4), inspired by RF3BP pulsating-rotating formulations.
GPU-Accelerated Brain Image Processing Pipeline for OpenNeuro Datasets
WACV 2026 RWS Challenge: Building object detectors that maintain consistent performance across seasons, weather patterns, and day-night cycles in thermal imagery.
A Brain-Computer Interface (BCI) system that visualizes memory formation patterns in real-time, helping users optimize learning and recall through neurofeedback.
bridging quantum computing and neural networks to unlock computational capabilities impossible with classical systems alone. Built for researchers, developers, and enterprises seeking quantum advantage in machine learning.
Add a description, image, and links to the framework-matplotlib topic page so that developers can more easily learn about it.
To associate your repository with the framework-matplotlib topic, visit your repo's landing page and select "manage topics."