Research & Education

samples-for-ai is a comprehensive collection of deep learning samples and projects designed to help beginners get started with deep learning. It encompasses a wide range of classic deep learning algorithms and applications, supporting multiple frameworks including TensorFlow, CNTK (BrainScript and Python), PyTorch, Caffe2, Keras, MXNet, Chainer, and Theano. The project offers samples in Visual Studio solution format, making it accessible for users leveraging Microsoft Visual Studio Tools for AI or Open Platform for AI. Users can run samples locally or submit jobs to OpenPAI, providing flexibility in deployment. This open-source initiative encourages contributions and adheres to the Microsoft Open Source Code of Conduct, fostering a collaborative environment for deep learning development.

The Global Index on Responsible AI is a multidimensional tool designed to ignite global action on responsible AI by providing local evidence. It measures progress in 138 countries and jurisdictions, uniquely using human rights-based benchmarks. The index offers comprehensive, comparable, and country-level data, allowing users to discover evidence, trends, rankings, and scoring. Its methodology is built on collaboration and consultation, encompassing three pillars, three dimensions, and 19 thematic areas, developed with stakeholders globally. The tool defines responsible AI through a human rights lens, making it a crucial resource for understanding and tracking AI development worldwide.

Speech-Separation-Paper-Tutorial is an invaluable resource for anyone interested in speech separation based on neural networks. This GitHub repository compiles a comprehensive collection of papers, models, and related resources spanning from 2016 to 2025. It offers detailed overviews, model timelines, and performance comparisons across various datasets like WSJ0-2Mix, WHAM!, and LibriMix. Users can explore different model categories, including deterministic vs. generative approaches, network architectures like dual-path and Conv-TasNet, and learning methods such as predictive and unsupervised techniques. The tutorial also delves into multi-modal speech separation, evaluation metrics like SI-SNRi and SDRi, and provides information on key datasets, making it a central hub for academic research and development in the field.

SUPIR is an open-source project dedicated to developing practical algorithms for photo-realistic image restoration in real-world scenarios. It provides advanced capabilities for enhancing image quality, including super-resolution and the ability to handle various degradations. The project emphasizes achieving high generalization and image quality, with options for both quality-oriented and fidelity-oriented settings. Users can choose between different model versions (SUPIR-v0Q and SUPIR-v0F) depending on their specific needs, such as general high quality or better detail preservation for light degradations. An online demo, SupPixel AI, is also available for easy access to its cutting-edge AI technology for image processing and upscaling.

PSLD is an AI research tool hosted on Hugging Face Spaces, intended for experimentation and research in artificial intelligence. While its purpose is to facilitate the exploration of AI algorithms, the current live version is encountering significant runtime errors, preventing it from functioning as intended. The tool's logs indicate issues with downloading stable diffusion pretrained weights and loading model checkpoints, suggesting it is in a developmental or unmaintained state. It aims to provide a platform for researchers and AI enthusiasts to test and develop AI models.

SupContrast offers a PyTorch implementation of "Supervised Contrastive Learning" and, incidentally, "A Simple Framework for Contrastive Learning of Visual Representations" (SimCLR). This repository serves as a reference, illustrating these methods using CIFAR datasets. It includes a `SupConLoss` function that takes features and labels, degenerating to SimCLR loss if labels are not provided. The implementation provides comparison results on CIFAR-10 and CIFAR-100, showcasing improved accuracy over standard cross-entropy. It also details running instructions for standard cross-entropy, supervised contrastive learning, and SimCLR, including pretraining and linear evaluation stages, and supports custom datasets.

(DEPRECIATED) Multi-Model Ensemble Deepfake Detection & Forensics is a tool that leverages an ensemble of AI models to analyze images and determine if they are AI-generated or real. It provides a comprehensive deepfake detection and forensics solution, offering detailed analysis and confidence scores from each contributing model. While the tool's core functionality is deepfake detection, its current status indicates it is deprecated and experiencing runtime errors, specifically a 'Repository Not Found' issue when trying to load a preprocessor configuration. This suggests the tool is currently non-functional and requires maintenance or updates to its underlying model dependencies.

3D-Adapter is an AI-powered tool designed to generate high-quality 3D models. Users can create detailed 3D visuals by providing descriptions or various inputs, making it accessible for a range of creative and technical applications. The tool leverages advanced AI to interpret user specifications and produce geometry-consistent multi-view diffusion, which is crucial for realistic 3D generation. Based on research papers, 3D-Adapter offers a demo to showcase its capabilities, making it a valuable resource for 3D artists, researchers, and anyone looking to efficiently produce 3D content without extensive manual modeling. It is available as a free-to-use application.

The National AI Institute for Adult Learning and Online Education (AI-ALOE) is a research institute funded by the National Science Foundation (NSF) and Accenture. Led by the Georgia Institute of Technology, AI-ALOE is dedicated to developing a transformative AI-based model for online adult learning. This model aims to simultaneously leverage AI to improve online adult learning and use online adult education to advance AI. The institute conducts use-inspired fundamental research into AI, grounded in theories of human cognition and learning, supported by large-scale data, and evaluated across various testbeds. AI-ALOE collaborates with partners in higher education and educational technology to make online learning more available, affordable, and achievable.

Distribution-is-all-you-need is an open-source GitHub repository offering a comprehensive tutorial on fundamental probability distributions crucial for deep learning researchers. The resource leverages Python libraries to illustrate various distributions, including Uniform, Bernoulli, Binomial, Categorical, Multinomial, Beta, Dirichlet, Gamma, Exponential, Gaussian, Normal, Chi-squared, and Student-t. It delves into concepts like conjugate distributions and their relevance in Bayesian probability theory, explaining how prior and posterior distributions relate. The tutorial provides code examples for each distribution, making it a practical guide for understanding the mathematical underpinnings of deep learning models.

Siwalu develops AI-based image recognition technology, primarily through mobile applications, to identify animal breeds. Their apps, including Dog Scanner, Cat Scanner, and Horse Scanner, allow users to quickly determine the breed of their pets or other animals by scanning images. This technology provides specific information about various characteristics and traits, offering a reliable statement about the breed within seconds, including mixed breeds. Siwalu aims to increase knowledge about global biodiversity through universal animal recognition. The platform has garnered over 26 million app downloads and identifies nearly 2 million animals per month, demonstrating its widespread adoption and utility.

China AI policy research 🤗 offers a comprehensive and curated collection of Chinese artificial intelligence policy papers and regulations. This tool is designed to assist researchers, policymakers, and academics in understanding the evolving landscape of AI policy within China and its international implications. Users can easily browse and filter the extensive collection by various categories, including Domestic policies, International collaborations, AI Safety regulations, and Data governance. This functionality allows for targeted research and efficient navigation through a large volume of policy documents, making it an invaluable resource for anyone interested in the current state and future trends of AI policy.

Conjecture is an AI research company focused on developing a new AI architecture called Cognitive Emulation. This innovative approach aims to ensure the controllable and safe development of advanced AI technology by building systems that emulate human reasoning processes. The company addresses challenges in AI safety such as unpredictable outputs, incoherent responses, inept systems, and uninterpretable inner workings. Cognitive Emulation seeks to revolutionize AI deployment by allowing AI to be built by component, trained for specific tasks, automate real workflows, and solve more complex problems, ultimately leading to more reliable and trustworthy AI. Conjecture emphasizes the importance of safety as AI capabilities scale, offering an alternative to the risks associated with current scaling practices.

The CrewAI Multiagent Research Tool is a free application designed to streamline the research process. Users can input a topic, and the tool will leverage a multi-agent system built with CrewAI to search the web for relevant information. It then fetches the main content from the top search results and generates concise markdown summaries. This tool is particularly useful for quickly gathering information and synthesizing it into an easily digestible format, making it an efficient solution for initial research phases or for generating quick overviews on various subjects.

The Newsroom specializes in creating provenance tools for journalism, focusing on making information traceable and verifiable. They are developing the first implementation of C2PA standards for text, allowing audiences to verify the origin of digital content. This process involves AI identifying claims from canonical sources, matching statements in articles to original claims, and creating tamper-evident manifests. Additionally, The Newsroom provides comprehensive AI training for newsrooms, covering fundamentals, workflow mapping, leadership guidance, and technical innovation, to help media organizations build AI capabilities and make informed decisions about AI adoption.

tidybot2 is an open-source project providing a holonomic mobile manipulator designed for robot learning. It includes comprehensive hardware designs and software components for building and operating the robot. The platform supports various tasks, from phone teleoperation and data collection to policy training and inference. Its holonomic base allows for independent and simultaneous control of planar degrees of freedom, simplifying complex mobile manipulation tasks. The project offers a simulation environment for testing the codebase without physical hardware and detailed guides for assembly, usage, and software setup, making it accessible for researchers and developers in the field of robotics.

Time-Series-Forecasting-and-Deep-Learning is a comprehensive, open-source GitHub repository dedicated to curating resources for time series forecasting and deep learning. It serves as a valuable hub for researchers, data scientists, and students seeking to explore the latest advancements in the field. The repository meticulously organizes research papers, including those from 2017 up to 2026, alongside benchmarks, applications like TimeGPT, and various datasets. Additionally, it provides links to relevant courses, blogs, and code libraries, making it an all-in-one reference for anyone involved in time series analysis and model development. The structured content, including a table of contents, allows for easy navigation through a vast collection of academic and practical materials.

tslearn is an open-source machine learning toolkit specifically designed for time series analysis in Python. It provides a wide array of functionalities for tasks such as clustering, classification, and regression of time series data. The toolkit supports various data preprocessing steps, including scaling and resampling, and offers different distance metrics like Dynamic Time Warping (DTW). tslearn is built to be compatible with scikit-learn's API, allowing users to leverage familiar utilities for hyper-parameter tuning and pipelines. It also includes features for calculating barycenters, performing early classification, and working with UCR datasets, making it a versatile tool for researchers and practitioners in the field.

Transformer-MM-Explainability is an official PyTorch implementation for Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers. This open-source project offers a novel method to visualize and understand the decision-making processes of any Transformer-based network. It includes practical examples for popular models such as DETR, VQA, CLIP, and LXMERT, making it a valuable resource for researchers and developers working with multi-modal and encoder-decoder architectures. The tool provides notebooks for easy experimentation and reproduction of results, with clear instructions for setting up environments and running examples on GPUs, including Colab support.

SoTA-Point-Cloud is a GitHub repository offering an extensive survey of deep learning techniques applied to 3D point clouds. Published in IEEE TPAMI 2020, this resource covers major tasks such as 3D shape classification, 3D object detection, and 3D point cloud segmentation. It provides comparative results on numerous publicly available datasets, including ModelNet, KITTI, and Semantic3D. The repository also offers insightful observations and outlines future research directions, making it an invaluable resource for researchers and practitioners in the field of 3D computer vision. The maintainers regularly update the page with new results and suggestions.

Woodpecker is an innovative, training-free method designed to correct hallucinations in Multimodal Large Language Models (MLLMs). Unlike existing studies that require retraining models, Woodpecker operates in a post-remedy manner, making it easily adaptable to various MLLMs. It functions through five distinct stages: key concept extraction, question formulation, visual knowledge validation, visual claim generation, and hallucination correction. This approach not only enhances the accuracy of generated text by aligning it with image content but also offers interpretability through its intermediate outputs. Woodpecker has demonstrated significant improvements in accuracy on benchmarks like POPE, making it a valuable tool for researchers and developers working with MLLMs.

write-you-a-vector-db is a comprehensive tutorial designed to guide users through the process of integrating vector capabilities into relational database systems. The tutorial is built upon modified versions of educational database systems, specifically CMU-DB's BusTub for the C++ variant and RisingLight for the upcoming Rust version. Users will learn to implement vector storage, vector expressions, and vector indexes. This resource is ideal for those looking to deepen their understanding of vector database implementation, offering practical, hands-on experience. The project is actively developed and encourages community participation through a dedicated Discord server.

The Yi series models are a collection of open-source large language models developed from scratch by 01.AI. These models are designed to be bilingual, trained on a 3T multilingual corpus, and excel in language understanding, commonsense reasoning, and reading comprehension. The Yi-34B-Chat model has demonstrated strong performance, ranking highly on leaderboards like AlpacaEval. The series includes both chat-optimized and base models, with options for different parameter sizes (6B, 9B, 34B) and context window lengths (up to 200K). Yi models are built on the Transformer architecture, similar to Llama, but are not derivatives, utilizing independently created training datasets and infrastructure. They are available for deployment via pip, Docker, conda-lock, and llama.cpp, and can be fine-tuned or quantized for specific needs.

Awesome-Adaptation-of-Agentic-AI is a curated repository featuring a comprehensive list of academic papers focused on the adaptation strategies of agentic AI systems. This resource is designed for researchers and practitioners interested in the evolving field of agentic AI, offering insights into various adaptation methods. The repository categorizes papers based on agent adaptation (tool execution signaled, agent output signaled) and tool adaptation (agent-agnostic, agent-supervised), detailing development timelines, methods, venues, tasks, tools, agent backbones, and tuning techniques. It serves as a valuable reference for understanding the latest advancements and research trends in making AI agents more adaptive and intelligent.