Research & Education

rnn-tutorial-rnnlm is an open-source project available on GitHub, offering a comprehensive tutorial for implementing Recurrent Neural Networks (RNNs). Specifically, it focuses on Part 2 of a tutorial series, guiding users through the process of building an RNN in Python and Theano. The repository includes all necessary code, a Jupyter Notebook for interactive learning, and detailed setup instructions. It covers both local development environments and advanced configurations for CUDA-enabled GPU instances on platforms like EC2, making it suitable for developers looking to understand and implement RNNs for language modeling and other sequential data tasks. The project is licensed under Apache-2.0.

CS231 is an open-source GitHub repository containing comprehensive solutions for the assignments of Stanford's renowned CS231n course, "Convolutional Neural Networks for Visual Recognition." Developed by cthorey, this resource is invaluable for students and researchers delving into deep learning and computer vision. The repository features practical implementations of core concepts, such as batch normalization, offering clear examples and code for understanding complex neural network architectures. Beyond the code, the creator has also published related blog posts, providing additional insights and explanations for the assignments. It serves as an excellent supplementary material for those studying the CS231n curriculum or anyone looking to deepen their understanding of convolutional neural networks through hands-on examples.

AILA is an AI-powered education technology platform designed to personalize learning experiences for students globally. It aims to transform traditional education by offering intelligent solutions that adapt to individual learning styles and paces. The platform rebuilds how schools operate, integrating personalized timetables, assignments, progress tracking, and performance insights into one comprehensive system. AILA maps and masters every skill, topic, and milestone through real-time insights, AI guidance, and smart recommendations. It caters to students, teachers, and lifelong learners, ensuring that learning is dynamic and responsive to individual needs, moving from confusion to comprehension faster.

Deep-Learning-Papers-Reading-Roadmap is a comprehensive GitHub repository designed to guide individuals eager to learn deep learning. It offers a structured reading roadmap, starting with historical and basic papers, then progressing to advanced methods and specific application areas. The roadmap is organized to move from outline to detail, old to state-of-the-art, and generic to specific topics, ensuring a logical learning path. It covers key areas such as Deep Learning History, ImageNet Evolution, Speech Recognition, various Deep Learning Methods (including optimization, unsupervised learning, RNNs, and reinforcement learning), and more. The repository is continuously updated with new and relevant papers, making it a valuable resource for continuous learning in the rapidly evolving field of deep learning.

LLMs-Zero-to-Hero is an open-source educational resource designed to guide individuals from basic understanding to advanced proficiency in Large Language Models (LLMs). The project emphasizes a hands-on approach, providing fully handwritten code examples and detailed explanations for each concept. It covers a wide range of topics, including the training process of dense and MOE models, pre-training, fine-tuning (SFT, DPO, RLHF), and deployment strategies like inference optimization and quantization. The resource also includes配套视频讲解 (accompanying video explanations) on Bilibili and offers GPU mirror images for model training, with a minimum requirement of 3090/4090 GPUs. It aims to provide a systematic learning path for aspiring LLM developers.

LLM-Dojo is a lightweight, open-source framework designed for post-training large language models (LLMs). It offers comprehensive support for various training methodologies, including Supervised Fine-Tuning (SFT), Reinforcement Learning from Human Feedback with Value Regularization (RLVR), On-Policy Knowledge Distillation (On-Policy KD), and Guide Knowledge Distillation (Guide KD). The platform also facilitates mixed training approaches, enabling single-round or multi-round Guide distillation, multi-teacher distillation, and reward mixed training. A key feature is its automated data shunting capabilities. Built on a refactored OpenRLHF core, LLM-Dojo streamlines the framework by retaining only the essential RLVR components and integrating advanced KD and Guide-KD techniques, making it suitable for rapid fine-tuning experiments with features like Deepspeed support, LoRA/QLoRA, and automatic chat template adaptation.

I, Saras is an AI-powered exam mentor specifically designed for students preparing for the UPSC exams in India. The platform provides a comprehensive ecosystem for learning, practicing, and staying updated, all within a unified AI environment. Users can chat with the AI mentor to resolve doubts instantly, receiving context-aware and topic-based explanations tailored to UPSC standards. It also offers AI-curated question sets, including Previous Year Questions (PYQs), with smart categorization, detailed explanations, and adaptive practice modes. Furthermore, I, Saras acts as a daily news companion, providing AI-curated current affairs with syllabus-linked insights and analysis to keep aspirants exam-ready. The tool aims to offer a smarter, faster, and calmer way to prepare, filtering out information overload and providing accurate, syllabus-aligned answers.

harmonic-oscillator-pinn offers an open-source code implementation for a physics-informed neural network (PINN) applied to a harmonic oscillator. This tool serves as a practical example for understanding and experimenting with PINNs, which integrate physical laws into neural network training. It is specifically designed to accompany a blog post by Ben Moseley, providing a hands-on resource for researchers and students interested in scientific machine learning and the application of AI to solve differential equations. The repository includes the necessary code to replicate the experiments and insights discussed in the associated blog post, making it a valuable educational and research asset.

mlhub123 is a curated online hub offering a vast collection of resources for machine learning and deep learning enthusiasts. It provides organized navigation to various categories including news, tools, community forums, quality blogs, resource retrieval, competitions, course learning, open-source books, and practical projects. The platform also features documentation for popular Python and C/C++ libraries and frameworks. It's an excellent starting point for anyone looking to explore or deepen their knowledge in AI, offering links to academic papers, datasets, and learning paths from renowned institutions and experts. The site aims to be a central point for discovering and utilizing machine learning resources.

ml4a.github.io serves as a comprehensive open-source resource dedicated to machine learning for artists. It offers a rich collection of notes, practical demonstrations, and interactive visuals designed to help creatives understand and apply machine learning concepts in their artistic endeavors. The platform also features various art projects that showcase the integration of ML, providing inspiration and tangible examples. This initiative aims to bridge the gap between complex machine learning technologies and the artistic community, making these tools accessible for creative exploration and innovation. It's a valuable hub for anyone interested in the intersection of AI and art.

pytorch-seq2seq offers comprehensive tutorials for understanding and implementing sequence-to-sequence (seq2seq) models using the PyTorch deep learning framework and TorchText library. The repository focuses on practical application, guiding users through the process of training models for neural machine translation, specifically from German to English. It covers foundational seq2seq concepts, including encoder-decoder models with LSTMs and GRUs, and delves into advanced topics like attention mechanisms to alleviate information compression problems. The tutorials are structured to build knowledge progressively, starting with basic workflows and moving to more sophisticated architectures. It also provides necessary setup instructions, including dependency installation and spaCy model downloads, making it a valuable resource for those looking to implement and experiment with seq2seq models.

Pollo AI is an all-in-one AI platform for generating high-quality videos and images. Users can transform text prompts, images, or videos into new creative content using a wide array of AI models like Pollo 2.5, Sora 2, and Stable Diffusion. The platform supports various video creation features, including AI Image to Video, Text to Video, Reference to Video, and AI Animation. For images, it offers AI Image Generator, Image to Image AI, and AI Photo Editor. Additionally, Pollo AI provides advanced editing tools like video upscalers, enhancers, object removers, and face swap capabilities, making it suitable for content creators, marketers, and businesses looking to produce engaging visual content efficiently.

aigc is an open-source electronic book titled "Unlocking the Potential of Large Language Models: Real-World Use Cases," focusing on the development and architectural design of LLM applications. Authored by Phodal and collaborators, this resource delves into the foundational knowledge of LLMs and their practical applications. It covers essential topics such as Prompt engineering, including writing, development, and management, as well as exploring the capabilities of advanced LLMs. The book also provides insights into LLM application development patterns and architectural designs, offering guidance on building custom models based on open-source solutions and implementing LLMOps. It serves as a comprehensive guide for understanding and implementing LLM-driven software development processes.

Awesome-Chinese-LLM is a comprehensive open-source repository dedicated to Chinese large language models (LLMs). The collection prioritizes models that are smaller in scale, suitable for private deployment, and have lower training costs, making them accessible to a wider range of users. It encompasses a variety of resources, including foundational base models like ChatGLM, LLaMA, Baichuan, and Qwen, as well as models fine-tuned for vertical domains such as healthcare, law, finance, and education. Beyond models, the repository also provides valuable datasets for pre-training, SFT, and preference alignment, along with tutorials covering LLM basics, prompt engineering, application development, and practical implementation. This makes it an invaluable resource for researchers, developers, and practitioners working with Chinese LLMs.

deep-learning-from-scratch-2 is a comprehensive support site for the book "Deep Learning from Scratch 2 ―Natural Language Processing Edition" (O'Reilly Japan, 2018). This GitHub repository offers all the source code used throughout the book, making it an invaluable resource for readers looking to implement and experiment with the concepts discussed. The repository is meticulously organized by chapter, with dedicated folders for each, alongside common utilities and dataset-related code. It requires Python 3.x, NumPy, and Matplotlib, with optional support for SciPy and CuPy. The code is released under an MIT license, allowing for free commercial and non-commercial use. Additionally, the site provides links to errata and contact information for reporting new errors, ensuring the accuracy and usability of the learning materials.

langchain-kr offers a comprehensive Korean tutorial for LangChain, built upon the official LangChain documentation, cookbooks, and practical examples. This resource is designed to help Korean speakers understand and utilize LangChain with greater ease and effectiveness. The tutorial covers a wide range of topics, from basic concepts and prompt engineering to advanced techniques like RAG, LangChain Expression Language (LCEL), and multi-agent collaboration with LangGraph. It includes practical examples, YouTube video explanations, and blog posts, making it a valuable learning resource for anyone looking to master LangChain in Korean. The project is open-source and encourages contributions from the community.

llms-from-scratch-cn is an open-source educational project by Datawhale, designed to help developers and researchers build large language models (LLMs) from the ground up. It offers a comprehensive learning path, emphasizing practical implementation over theoretical concepts. The project focuses on understanding LLM architecture, providing step-by-step tutorials to construct models such as GLM4, Llama3, and RWKV6. It includes detailed code examples, covering encoding, pre-training, and fine-tuning processes, making it accessible for individuals with basic Python and PyTorch knowledge to delve deep into LLM principles.

Starter-Guide, developed by the PKU-DAIR team, is an open-source repository designed to provide a comprehensive guide for beginners in the fields of data management (DM) and artificial intelligence (AI). It consolidates core papers and shared experiences from the team to help newcomers quickly familiarize themselves with cutting-edge areas and build a solid technical foundation. The guide covers various research directions including AI systems, AutoML, Database, AI Agent, Data-Centric ML, Diffusion Models, AI for Science, and Graph. It aims to support users in their learning and research journeys, whether they are just starting out or looking to deepen their understanding.

Prompt-Engineering-Guide-Chinese is a comprehensive, open-source guide designed to help individuals understand and master prompt engineering. It is a translated and updated version of a popular English guide, specifically enhanced with AIGC (AI-Generated Content) prompt sections to make the learning process more accessible for Chinese-speaking users. The guide covers the development and optimization of prompts for effectively utilizing large language models (LLMs) across various applications and research topics. It aims to improve understanding of LLMs' capabilities and limitations, offering insights for researchers to enhance LLMs' performance on tasks like Q&A and arithmetic reasoning, and for developers to design powerful prompting techniques for LLM interfaces.

Gurully is an AI-powered online platform designed to help individuals prepare for major English proficiency exams including PTE, IELTS, Duolingo, and CELPIP. The platform offers full-length mock tests with instant AI scoring, providing detailed feedback on strengths and areas for improvement. Users can access one free mock test for each exam type to familiarize themselves with the format and identify their current skill level. Beyond mock tests, Gurully provides sample answers, predictive questions, and flexible pricing plans, making it a comprehensive resource for English language learners aiming to achieve high scores for study or immigration purposes. It also features a mobile app for both Android and iOS for convenient practice on the go.

Build-A-Large-Language-Model-CN is an open-source project on GitHub that offers a Chinese translation of the e-book "Build a Large Language Model (From Scratch)". This resource is designed for learners eager to delve into the core principles and practical implementation of large language models (LLMs), including architectures like GPT, their training processes, and application development. The project aims to make this valuable educational material accessible to a wider Chinese-speaking audience. It includes the translated Chinese version, the original English e-book, and all translated images. The author also provides personal insights and interpretations to enhance understanding, alongside practical code examples for hands-on learning.

Depth of ML - GATE DA is an e-learning platform dedicated to providing comprehensive and in-depth content for students preparing for the GATE Data Science and Artificial Intelligence exam. The platform emphasizes hands-on learning to ensure a thorough understanding of the subject matter. Founded by IISc graduates Deval and Saloni, Depth of ML aims to deliver quality education tailored specifically for GATE DA aspirants. While the platform will stop accepting new subscriptions after January 10, 2026, existing active subscriptions will continue to receive full support, including doubt resolution from the founders, ensuring no compromise on commitments.

ZenseAI is an AI-powered education platform designed to transform teaching and learning experiences for schools. It offers a comprehensive suite of tools for both educators and students, including access to multiple AI models for diverse educational tasks, efficient organization and management of educational resources, and personalized AI tutors tailored to individual learning needs. The platform also provides AI-powered exercises with instant progress tracking, tools for assigning missions, monitoring student progress, and detecting dangerous behavior. ZenseAI aims to integrate AI seamlessly into the classroom, supporting digital transformation and offering services like STEAM courses, teacher training, and government funding assistance.

how-to-optim-algorithm-in-cuda is a comprehensive open-source repository dedicated to optimizing algorithms using CUDA. It offers a wealth of resources including code implementations for fundamental CUDA operators like reduce, softmax, and elementwise operations, as well as detailed learning notes and blog translations related to GPU and large language models. The project covers advanced topics such as CUTLASS, CuTe DSL, Triton, and PTX ISA, making it an invaluable learning tool for developers aiming to enhance the performance of their CUDA code. It also includes notes on large language model inference/training optimization and GPU/AI system papers.