Research & Education

BAIOSPHERE is the Bavarian AI Network, dedicated to connecting researchers, companies, and institutions to build a robust and trustworthy AI future. It serves as a central hub for AI innovation in Bavaria, focusing on strengthening technological sovereignty through initiatives like the Bavarian AI Foundation Model. The network facilitates knowledge exchange, promotes AI research and development, and highlights funding opportunities. BAIOSPHERE also organizes events like the BAIOSPHERE Health Track and Munich AI Lectures, bringing together leading AI experts to discuss the future of artificial intelligence and its applications across various sectors.

The Minor in AI program by IIT Ropar is an educational offering designed to equip students with comprehensive knowledge and practical skills in artificial intelligence. This course delves into both fundamental and advanced concepts within AI, encompassing critical areas such as machine learning, data science, and neural networks. The curriculum is structured to provide a robust understanding of AI principles and their applications, preparing students for the demands of AI-driven industries. By focusing on core AI disciplines, the program aims to foster expertise and innovation among its participants, enabling them to contribute effectively to the rapidly evolving field of artificial intelligence.

Core Defender AI focuses on empowering the future by building AI and Quantum-ready platforms across various sectors. For education, their AIR for Kids program provides an interactive platform for children aged 6-14 to learn AI by building chatbots and programming robots. In healthcare, they offer a modular AI assistant framework, exemplified by NOVA for Ayumetrix, designed for secure, agentic AI systems. For local businesses, Core Defender provides AI companions that handle bookings, FAQs, and lead generation for restaurants, clinics, and retail. A core differentiator is their commitment to security, which is built into every platform from the foundation, and their quantum-ready architecture ensures systems evolve with future computing capabilities.

ml-retreat is a comprehensive, open-source machine learning journal designed for individuals with intermediate to advanced knowledge in the field. It functions as a personal learning repository, offering in-depth explanations of fundamental concepts alongside curated resources for more nuanced subjects. The journal currently focuses on mechanistic interpretability, providing detailed notes, recommended readings, and watchlists from prominent figures like Ilya Sutskever and Andrej Karpathy. It covers topics such as Large Language Models, Graph Neural Networks, and AlphaFold 3, making it an invaluable resource for structured self-study and continuous learning in cutting-edge AI.

Axis India Machine Learning provides comprehensive courses and mentorship programs designed to educate individuals in the fields of machine learning, artificial intelligence, and data science. The platform's core mission is to teach users "How To Learn," emphasizing practical skills and understanding over rote memorization. While specific course details are not extensively listed, the offering of "Introduction to Full Stack Generative AI" suggests a focus on cutting-edge and in-demand AI technologies. The platform appears to be an educational resource for those looking to develop or enhance their expertise in AI and related domains.

Lora Finetuning Guide is an educational resource hosted on Hugging Face Spaces, designed to help users understand and implement LoRA (Low-Rank Adaptation) finetuning. This guide enables individuals to fine-tune generative AI models, such as Stable Diffusion, to integrate specific concepts. Users can provide their own images and a corresponding dataset description to customize a model, resulting in a personalized AI model that has learned the desired concept. It serves as a practical educational tool for those interested in customizing AI models and exploring advanced machine learning techniques.

Breyta is an AI tool designed to empower users to build sophisticated workflows using coding agents. It allows for the creation of custom AI agents that can automate various development tasks, streamlining the software development lifecycle. The platform focuses on integrating these AI agents seamlessly into existing systems and workflows, enhancing productivity and efficiency. Breyta aims to provide a flexible environment for developers and teams to leverage AI for code generation, testing, and other programming-related activities, ultimately accelerating project delivery and innovation.

Review My eLearning is the first AI-integrated eLearning review platform designed to enhance and streamline the course feedback and quality assurance processes. It offers compatibility with all major eLearning development tools, including Articulate Storyline 360, Rise, and Captivate. By leveraging cutting-edge AI insights, the platform helps users efficiently review their eLearning content, identify areas for improvement, and ensure high-quality educational materials. This tool is ideal for professionals involved in course creation and development who seek to optimize their review workflows and integrate advanced AI capabilities into their feedback loops.

The deep-learning-illustrated repository on GitHub offers the complete code and Jupyter notebooks that complement the 'Deep Learning Illustrated' book by Jon Krohn, Grant Beyleveld, and Aglaé Bassens. This resource provides a visual and interactive approach to understanding artificial neural networks and deep learning. It covers a wide range of topics from biological and machine vision to natural language processing, generative adversarial networks, and deep reinforcement learning. Users can find step-by-step installation guides and all code examples, making it suitable for those seeking a practical introduction to AI and deep learning implementation. The notebooks are primarily in TensorFlow, with notes on converting to TensorFlow 2.x.

MusicGPT is an innovative application designed for generating music from natural language prompts. It leverages Large Language Models (LLMs) that run locally, ensuring performant music creation across different platforms without the need for extensive dependencies like Python or complex machine learning frameworks. Currently, it supports MusicGen by Meta, with plans to integrate more music generation models. Users can interact with MusicGPT through a chat-like UI mode, which stores chat history, allows playing generated samples, and generates music in the background. Alternatively, a CLI mode enables direct music generation and playback in the terminal, with configurable sample lengths. It offers flexibility in model selection and GPU usage, though powerful hardware is recommended for larger models.

Welcome to EdgeAI for Beginners, a comprehensive course that bridges the gap between powerful AI capabilities and practical, real-world deployment on edge devices. This course empowers users to harness AI's potential directly where data is generated and decisions need to be made. It covers fundamental concepts, popular models like SLMs (Phi-4, Mistral-7B, Gemma), inference techniques, device-specific applications, and model optimization. The curriculum also delves into the development of intelligent Edge AI agents, production deployment strategies, and includes hands-on workshops to build local AI chat applications, RAG pipelines, and multi-agent orchestration systems. The course is designed for a wide range of learners, from beginners to experts, with modules covering various aspects of Edge AI.

Morpheus Uncensored Tts is a text-to-speech tool available as a Hugging Face Space, allowing users to generate natural-sounding speech from text input. A key feature is the ability to add emotive tags like <laugh> or <sigh> to the text, which helps in creating more human-like and expressive audio outputs. This tool is particularly useful for content creators looking to add dynamic voiceovers or experiment with uncensored audio generation. The application provides an audio output that can be listened to directly, making it suitable for quick prototyping and experimentation in voice synthesis.

Efficient-LLMs-Survey is a comprehensive academic survey focusing on efficient large language models (LLMs), published in Transactions on Machine Learning Research (TMLR) in May 2024. This resource systematically reviews techniques and methods for improving LLM efficiency, addressing the substantial resource demands of these powerful models. The survey categorizes the literature into three main areas: model-centric methods (e.g., compression, quantization, pruning), data-centric methods (e.g., data selection, prompt engineering), and framework-centric perspectives (e.g., system-level optimization, LLM frameworks). It serves as a valuable resource for researchers and practitioners seeking to understand the current landscape of efficient LLM research and inspire future contributions to the field.

kDimensions offers a visual introduction to deep learning and artificial intelligence, aiming to simplify complex concepts for learners. The platform highlights a book, "A Visual Introduction to Deep Learning," which is praised for its no-frills approach and ability to build visual intuition about deep learning and neural networks. This resource is designed to combat information overload in the rapidly evolving field of AI, making it accessible for individuals looking to understand the underpinnings of neural networks and explore career opportunities in AI. The book is recommended by experts for its clear explanations and effective visual aids.

Mathos AI (also known as MathGPTPro) is a comprehensive AI math solver, calculator, and tutor designed to assist over 5 million students with subjects like calculus, algebra, physics, and engineering. It provides instant, accurate solutions with detailed step-by-step explanations, claiming 20% higher accuracy than ChatGPT. Users can input equations by typing, screenshotting, using formulas, or voice, and the tool recognizes handwriting and complex formulas. Beyond problem-solving, Mathos AI offers an interactive learning experience with features like an advanced graphing calculator, PDF homework helper for chatting with documents, and personalized AI tutoring that adapts to individual learning styles.

Mistral 7B Instruct GGUF Run On CPU Basic is a Hugging Face Space that provides a user-friendly interface to interact with the Mistral 7B Instruct model. This tool is designed for basic text generation on a CPU, making it accessible for experimentation and personal projects without requiring high-end GPUs. Users can input messages and receive AI-generated responses, with options to fine-tune the output's randomness (temperature) and focus (top_p) using intuitive sliders. It functions as a general assistant, capable of various conversational tasks.

The EPSRC National Edge Artificial Intelligence Hub is dedicated to world-class fundamental research focused on protecting the quality of data and learning associated with AI algorithms, particularly when subjected to cyber attacks within Edge Computing environments. The hub offers solutions and support for the entire Edge AI ecosystem, fostering engagement, education, collaboration, and innovation across various industries. It aims to empower stakeholders to harness the full potential of edge AI through initiatives like the Edge AI Engage, Educate, Connect, and Incubate programs. The hub also facilitates research through various workstreams, including cyberdisturbance modelling, AI-driven edge data guard, and quantum machine learning, and provides opportunities for academic and industrial partnerships.

MUST Research Labs specializes in data science, cognitive computing, artificial intelligence, machine learning, and advanced analytics. They provide a comprehensive suite of services including research, certification programs, training, education, and tutorials. The organization also offers brainstorming sessions, consultancy, and development services, alongside solutions and products in these advanced technological fields. Their offerings cater to various needs, from academic research and professional development to practical application and product creation, aiming to foster a robust ecosystem around data science and AI.

Happy-LLM offers a comprehensive, open-source tutorial designed to help users deeply understand the principles and training processes of large language models (LLMs). Starting with fundamental NLP concepts, the project progressively delves into LLM architecture, covering Transformer mechanisms, pre-trained language models, and the core ideas behind LLMs. It provides practical guidance on building a complete LLaMA2 model from scratch, including tokenizer training and pre-training small LLMs. The tutorial also covers advanced topics like supervised fine-tuning, efficient fine-tuning methods (LoRA/QLoRA), and real-world applications such as RAG and Agent technologies, making it ideal for those looking to gain hands-on experience in the LLM field.

ner-lstm is an open-source project that provides an implementation of Named Entity Recognition (NER) using multilayered bidirectional Long Short-Term Memory (LSTM) networks. This tool is based on the approach described in a research paper published at the ICON-16 conference. It leverages TensorFlow for its deep learning architecture and supports classification tasks for named entities in text corpora. The project includes functionalities for generating embedding models (Word2Vec, GloVe, RnnVec), preparing input data by resizing datasets and converting sentences to embeddings, and running the deep neural network. It has been tested on CoNNL 2003 NER Shared Task and the ICON-2013 Hindi NER dataset, demonstrating its applicability to both English and Hindi languages. The code is available on GitHub, making it accessible for developers and researchers interested in natural language processing.

LLM-eval-survey is the official GitHub page for the survey paper "A Survey on Evaluation of Large Language Models." It functions as a central repository for researchers and practitioners, offering a curated collection of papers and resources focused on the evaluation of large language models (LLMs). The repository organizes papers by various evaluation aspects, including natural language processing tasks (understanding, sentiment analysis, text classification, inference, reasoning, generation, summarization, dialogue, translation, question answering), robustness, ethics, biases, trustworthiness, and applications in social science, natural science, engineering, medical, and agent domains. It also provides updates on new paper versions and welcomes community contributions.

llm-paper-daily is an open-source GitHub repository dedicated to curating and updating a daily list of research papers focused on large language models (LLMs). This resource is invaluable for researchers, academics, and enthusiasts who need to stay abreast of the rapid advancements in the LLM field. The repository offers a structured way to discover new papers, often including summaries and direct links to the original sources, making it easier to navigate the vast amount of new research. By providing a centralized and frequently updated collection, llm-paper-daily significantly reduces the time and effort required to track the latest developments in AI research, particularly in areas like agents, RAG (Retrieval-Augmented Generation), and general LLM applications.

LLMs-from-scratch is an open-source GitHub repository that provides comprehensive code for developing, pretraining, and finetuning GPT-like Large Language Models (LLMs) from the ground up using PyTorch. This repository is the official code companion for the book "Build a Large Language Model (From Scratch)", guiding users step-by-step through the process of creating their own functional LLM. It includes code for various stages, from understanding attention mechanisms to finetuning for text classification and instruction following. The project emphasizes a hands-on approach, allowing users to implement LLM components without relying on external LLM libraries, making it an invaluable resource for deep learning practitioners and researchers.

llm-books is an open-source GitHub repository offering a comprehensive collection of practical notes and code examples for developing applications with Large Language Models (LLMs). It serves as an educational resource for developers looking to understand and implement LLM technologies. The repository covers a wide range of topics, including an overview of LLMs, hands-on chatbot development using the OpenAI API, an introduction to LangChain modules (Chains, Agents, Callbacks), embedding techniques, and building enterprise knowledge bases with LlamaIndex. It also delves into advanced subjects like HuggingGPT, LLMOps, Agent systems, RAG (Retrieval-Augmented Generation) strategies, and LLM application evaluation. The resource includes insights into domestic model vendor APIs and provides a structured learning path for AI application development.