LLM-Synthetic-Data is a repository focused on real-time, fine-grained LLM-Synthetic-Data generation. It includes methods, surveys, and application areas related to synthetic data for language models. The repository covers topics like pre-training, instruction tuning, model collapse, LLM benchmarking, evaluation, and distillation. It also explores application areas such as mathematical reasoning, code generation, text-to-SQL, alignment, reward modeling, long context, weak-to-strong generalization, agent and tool use, vision and language, factuality, federated learning, generative design, and safety.

Awesome LLM Strawberry is a collection of research papers and blogs related to OpenAI Strawberry(o1) and Reasoning. The repository is continuously updated to track the frontier of LLM Reasoning.

EDA-AI is a repository containing implementations of cutting-edge research papers in the field of chip design. It includes DeepPlace, PRNet, HubRouter, and PreRoutGNN models for tasks such as placement, routing, timing prediction, and global routing. Researchers and practitioners can leverage these implementations to explore advanced techniques in chip design.

This repo contains official implementations of **Coach AI Badminton Project** from Advanced Database System Laboratory, National Yang Ming Chiao Tung University supervised by Prof. Wen-Chih Peng. The high-level concepts of each project are as follows: 1. Visualization Platform published at _Physical Education Journal 2020_ aims to construct a platform that can be used to illustrate the data from matches. 2. Shot Influence and Extension Work published at _ICDM-21_ and _ACM TIST 2022_ , respectively introduce a framework with a shot encoder, a pattern extractor, and a rally encoder to capture long short-term dependencies for evaluating players' performance of each shot. 3. Stroke Forecasting published at _AAAI-22_ proposes the first stroke forecasting task to predict the future strokes of both players based on the given strokes by ShuttleNet, a position-aware fusion of rally progress and player styles framework. 4. Strategic Environment published at _AAAI-23 Student Abstract_ designs a safe and reproducible badminton environment for turn-based sports, which simulates rallies with different angles of view and designs the states, actions, and training procedures. 5. Movement Forecasting published at _AAAI-23_ proposes the first movement forecasting task, which contains not only the goal of stroke forecasting but also the movement of players, by DyMF, a novel dynamic graphs and hierarchical fusion model based on the proposed player movements (PM) graphs. 6. CoachAI-Challenge-IJCAI2023 is a badminton challenge (CC4) hosted at _IJCAI-23_. Please find the website for more details. 7. ShuttleSet published at _KDD-23_ is the largest badminton singles dataset with stroke-level records. - An extension dataset ShuttleSet22 published at _IJCAI-24 Demo & IJCAI-23 IT4PSS Workshop_ is also released. 8. CoachAI Badminton Environment published at _AAAI-24 Student Abstract and Demo, DSAI4Sports @ KDD 2023_ is a reinforcement learning (RL) environment tailored for AI-driven sports analytics, offering: i) Realistic opponent simulation for RL training; ii) Visualizations for evaluation; and iii) Performance benchmarks for assessing agent capabilities.

FedLLM-Bench is a realistic benchmark for the Federated Learning of Large Language Models community. It includes datasets for federated instruction tuning and preference alignment tasks, exhibiting diversities in language, quality, quantity, instruction, sequence length, embedding, and preference. The repository provides training scripts and code for open-ended evaluation, aiming to facilitate research and development in federated learning of large language models.

This repository, RAM, focuses on developing advanced algorithms and methods for Reasoning, Alignment, Memory. It contains projects related to these areas and is maintained by a team of individuals. The repository is licensed under the MIT License.

This repository is dedicated to the pruning of large language models (LLMs). It aims to serve as a comprehensive resource for researchers and practitioners interested in the efficient reduction of model size while maintaining or enhancing performance. The repository contains various papers, summaries, and links related to different pruning approaches for LLMs, along with author information and publication details. It covers a wide range of topics such as structured pruning, unstructured pruning, semi-structured pruning, and benchmarking methods. Researchers and practitioners can explore different pruning techniques, understand their implications, and access relevant resources for further study and implementation.

AI Hospital is a research repository focusing on the interactive evaluation and collaboration of Large Language Models (LLMs) as intern doctors for clinical diagnosis. The repository includes a simulation module tailored for various medical roles, introduces the Multi-View Medical Evaluation (MVME) Benchmark, provides dialog history documents of LLMs, replication instructions, performance evaluation, and guidance for creating intern doctor agents. The collaborative diagnosis with LLMs emphasizes dispute resolution. The study was authored by Zhihao Fan, Jialong Tang, Wei Chen, Siyuan Wang, Zhongyu Wei, Jun Xie, Fei Huang, and Jingren Zhou.

ChatWise is an offline tool that supports various AI models such as OpenAI, Anthropic, Google AI, Groq, and Ollama. It is multi-modal, allowing text-to-speech powered by OpenAI and ElevenLabs. The tool supports text files, PDFs, audio, and images across different models. ChatWise is currently available for macOS (Apple Silicon & Intel) with Windows support coming soon.

The Medical_Image_Analysis repository focuses on X-ray image-based medical report generation using large language models. It provides pre-trained models and benchmarks for CheXpert Plus dataset, context sample retrieval for X-ray report generation, and pre-training on high-definition X-ray images. The goal is to enhance diagnostic accuracy and reduce patient wait times by improving X-ray report generation through advanced AI techniques.

ebook2audiobook is a CPU/GPU converter tool that converts eBooks to audiobooks with chapters and metadata using tools like Calibre, ffmpeg, XTTSv2, and Fairseq. It supports voice cloning and a wide range of languages. The tool is designed to run on 4GB RAM and provides a new v2.0 Web GUI interface for user-friendly interaction. Users can convert eBooks to text format, split eBooks into chapters, and utilize high-quality text-to-speech functionalities. Supported languages include Arabic, Chinese, English, French, German, Hindi, and many more. The tool can be used for legal, non-DRM eBooks only and should be used responsibly in compliance with applicable laws.

Agents 2.0 is a framework for training language agents using symbolic learning, inspired by connectionist learning for neural nets. It implements main components of connectionist learning like back-propagation and gradient-based weight update in the context of agent training using language-based loss, gradients, and weights. The framework supports optimizing multi-agent systems and allows multiple agents to take actions in one node.

OpenNARS-for-Applications is an implementation of a Non-Axiomatic Reasoning System, a general-purpose reasoner that adapts under the Assumption of Insufficient Knowledge and Resources. The system combines the logic and conceptual ideas of OpenNARS, event handling and procedure learning capabilities of ANSNA and 20NAR1, and the control model from ALANN. It is written in C, offers improved reasoning performance, and has been compared with Reinforcement Learning and means-end reasoning approaches. The system has been used in real-world applications such as assisting first responders, real-time traffic surveillance, and experiments with autonomous robots. It has been developed with a pragmatic mindset focusing on effective implementation of existing theory.

MyScaleDB is a SQL vector database optimized for AI applications, enabling developers to manage and process massive volumes of data efficiently. It offers fast and powerful vector search, filtered search, and SQL-vector join queries, making it fully SQL-compatible. MyScaleDB provides unmatched performance and scalability by leveraging cutting-edge OLAP database architecture and advanced vector algorithms. It is production-ready for AI applications, supporting structured data, text, vector, JSON, geospatial, and time-series data. MyScale Cloud offers fully-managed MyScaleDB with premium features on billion-scale data, making it cost-effective and simpler to use compared to specialized vector databases. Built on top of ClickHouse, MyScaleDB combines structured and vector search efficiently, ensuring high accuracy and performance in filtered search operations.

MyScaleDB is a SQL vector database designed for scalable AI applications, enabling developers to efficiently manage and process massive volumes of data using familiar SQL. It offers fast and efficient vector search, filtered search, and SQL-vector join queries. MyScaleDB is fully SQL-compatible and production-ready for AI applications, providing unmatched performance and scalability through cutting-edge OLAP architecture and advanced vector algorithms. Built on top of ClickHouse, it combines structured and vectorized data management for high accuracy and speed in filtered searches.

The Alignment Handbook provides robust training recipes for continuing pretraining and aligning language models with human and AI preferences. It includes techniques such as continued pretraining, supervised fine-tuning, reward modeling, rejection sampling, and direct preference optimization (DPO). The handbook aims to fill the gap in public resources on training these models, collecting data, and measuring metrics for optimal downstream performance.

Xwin-LM is a powerful and stable open-source tool for aligning large language models, offering various alignment technologies like supervised fine-tuning, reward models, reject sampling, and reinforcement learning from human feedback. It has achieved top rankings in benchmarks like AlpacaEval and surpassed GPT-4. The tool is continuously updated with new models and features.

The repository 'Awesome-LLM-Preference-Learning' is the official repository of a survey paper titled 'Towards a Unified View of Preference Learning for Large Language Models: A Survey'. It contains a curated list of papers related to preference learning for Large Language Models (LLMs). The repository covers various aspects of preference learning, including on-policy and off-policy methods, feedback mechanisms, reward models, algorithms, evaluation techniques, and more. The papers included in the repository explore different approaches to aligning LLMs with human preferences, improving mathematical reasoning in LLMs, enhancing code generation, and optimizing language model performance.

LLM-Synthetic-Data is a repository focused on real-time, fine-grained LLM-Synthetic-Data generation. It includes methods, surveys, and application areas related to synthetic data for language models. The repository covers topics like pre-training, instruction tuning, model collapse, LLM benchmarking, evaluation, and distillation. It also explores application areas such as mathematical reasoning, code generation, text-to-SQL, alignment, reward modeling, long context, weak-to-strong generalization, agent and tool use, vision and language, factuality, federated learning, generative design, and safety.

DeepEval is a simple-to-use, open-source LLM evaluation framework specialized for unit testing LLM outputs. It incorporates various metrics such as G-Eval, hallucination, answer relevancy, RAGAS, etc., and runs locally on your machine for evaluation. It provides a wide range of ready-to-use evaluation metrics, allows for creating custom metrics, integrates with any CI/CD environment, and enables benchmarking LLMs on popular benchmarks. DeepEval is designed for evaluating RAG and fine-tuning applications, helping users optimize hyperparameters, prevent prompt drifting, and transition from OpenAI to hosting their own Llama2 with confidence.

Bench is a tool for evaluating LLMs for production use cases. It provides a standardized workflow for LLM evaluation with a common interface across tasks and use cases. Bench can be used to test whether open source LLMs can do as well as the top closed-source LLM API providers on specific data, and to translate the rankings on LLM leaderboards and benchmarks into scores that are relevant for actual use cases.

TrustLLM is a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, evaluation, and analysis of trustworthiness for mainstream LLMs, and discussion of open challenges and future directions. Specifically, we first propose a set of principles for trustworthy LLMs that span eight different dimensions. Based on these principles, we further establish a benchmark across six dimensions including truthfulness, safety, fairness, robustness, privacy, and machine ethics. We then present a study evaluating 16 mainstream LLMs in TrustLLM, consisting of over 30 datasets. The document explains how to use the trustllm python package to help you assess the performance of your LLM in trustworthiness more quickly. For more details about TrustLLM, please refer to project website.

LLM AutoEval is a tool that simplifies the process of evaluating Large Language Models (LLMs) using a convenient Colab notebook. It automates the setup and execution of evaluations using RunPod, allowing users to customize evaluation parameters and generate summaries that can be uploaded to GitHub Gist for easy sharing and reference. LLM AutoEval supports various benchmark suites, including Nous, Lighteval, and Open LLM, enabling users to compare their results with existing models and leaderboards.

Weave is a toolkit for developing Generative AI applications, built by Weights & Biases. With Weave, you can log and debug language model inputs, outputs, and traces; build rigorous, apples-to-apples evaluations for language model use cases; and organize all the information generated across the LLM workflow, from experimentation to evaluations to production. Weave aims to bring rigor, best-practices, and composability to the inherently experimental process of developing Generative AI software, without introducing cognitive overhead.

LLMStack is a no-code platform for building generative AI agents, workflows, and chatbots. It allows users to connect their own data, internal tools, and GPT-powered models without any coding experience. LLMStack can be deployed to the cloud or on-premise and can be accessed via HTTP API or triggered from Slack or Discord.

The VisionCraft API is a free API for using over 100 different AI models. From images to sound.

Kaito is an operator that automates the AI/ML inference model deployment in a Kubernetes cluster. It manages large model files using container images, avoids tuning deployment parameters to fit GPU hardware by providing preset configurations, auto-provisions GPU nodes based on model requirements, and hosts large model images in the public Microsoft Container Registry (MCR) if the license allows. Using Kaito, the workflow of onboarding large AI inference models in Kubernetes is largely simplified.

PyRIT is an open access automation framework designed to empower security professionals and ML engineers to red team foundation models and their applications. It automates AI Red Teaming tasks to allow operators to focus on more complicated and time-consuming tasks and can also identify security harms such as misuse (e.g., malware generation, jailbreaking), and privacy harms (e.g., identity theft). The goal is to allow researchers to have a baseline of how well their model and entire inference pipeline is doing against different harm categories and to be able to compare that baseline to future iterations of their model. This allows them to have empirical data on how well their model is doing today, and detect any degradation of performance based on future improvements.

Tabby is a self-hosted AI coding assistant, offering an open-source and on-premises alternative to GitHub Copilot. It boasts several key features: * Self-contained, with no need for a DBMS or cloud service. * OpenAPI interface, easy to integrate with existing infrastructure (e.g Cloud IDE). * Supports consumer-grade GPUs.

SPEAR (Simulator for Photorealistic Embodied AI Research) is a powerful tool for training embodied agents. It features 300 unique virtual indoor environments with 2,566 unique rooms and 17,234 unique objects that can be manipulated individually. Each environment is designed by a professional artist and features detailed geometry, photorealistic materials, and a unique floor plan and object layout. SPEAR is implemented as Unreal Engine assets and provides an OpenAI Gym interface for interacting with the environments via Python.

Magick is a groundbreaking visual AIDE (Artificial Intelligence Development Environment) for no-code data pipelines and multimodal agents. Magick can connect to other services and comes with nodes and templates well-suited for intelligent agents, chatbots, complex reasoning systems and realistic characters.