Research & Education

SWE-RL is an official codebase for "Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution," designed to scale reinforcement learning-based LLM reasoning for real-world software engineering tasks. It leverages open-source software evolution data and rule-based rewards to improve LLM performance. The codebase includes prompt templates and a flexible reward function API that supports various editing formats, including sequence similarity for search/replace changes and unified diffs. Additionally, SWE-RL features an Agentless Mini component for fast asynchronous inference, code refactoring, file-level localization, and repair, supporting OpenAI-compatible endpoints and Hugging Face models like Llama-3.3-70B-Instruct.

Deix S.r.l. specializes in developing innovative algorithms and applications by leveraging expertise in mathematical modeling, artificial intelligence, and optimization. They provide solutions that enable companies to make informed decisions and identify new business opportunities. Deix offers both ready-to-use products and tailor-made solutions designed to meet specific business needs. Their approach integrates internal knowledge and data to deliver high-quality, efficient results, as evidenced by client testimonials highlighting speed, technical expertise, and proactivity in solving complex challenges.

sqlite-vss is a SQLite extension designed to bring vector search capabilities directly into SQLite databases, leveraging the Faiss library for efficiency. It enables developers to build semantic search engines, recommendation systems, and question-and-answering tools by storing and querying vector embeddings. While not actively developed, with efforts now focused on sqlite-vec, it offers a robust solution for integrating vector search into applications using SQLite. Users can create virtual tables to store high-dimensional embeddings and perform k-nearest neighbor searches. It supports various languages through bindings like Python, Node.js, Deno, Ruby, Elixir, Go, and Rust, making it accessible to a wide range of developers.

Falcondale specializes in developing applied quantum machine learning and optimization solutions designed to deliver real-world impact. The company focuses on leveraging quantum intelligence to solve complex problems across various industries. Falcondale aims to provide a competitive edge through its advanced quantum technologies, offering solutions that go beyond traditional computational methods. Their expertise lies in translating cutting-edge quantum research into practical, deployable applications for businesses and organizations seeking innovative data analysis and optimization capabilities.

StableDiffusion-CheatSheet is an open-source resource designed to assist users in exploring and utilizing Stable Diffusion styles. It functions as a personal cheat sheet, offering a vast collection of over 833 manually tested styles, complete with notes for offline access. Users can easily copy style prompts with a single click and leverage robust search and filter functionalities to find specific artists or styles. The tool also allows for checking image metadata without needing to launch Stable Diffusion, simply by dragging and dropping images. Additionally, it provides extra notes on art styles and a simple way to calculate image dimensions. A 'just the data' version is available for those who prefer information without preview images, including artist details, categories, and a list of artists checked but unknown to Stable Diffusion.

Flashwise is an AI-powered education application designed to help students master any subject effortlessly. It leverages advanced AI models to create beautifully crafted flashcards tailored to individual learning needs, generating study sets in seconds. The app incorporates a scientifically-proven spaced repetition technique, intelligently tracking progress and adjusting flashcard prompts to review difficult concepts more often and space out mastered ones, ensuring long-term retention. Flashwise also features an AI bot for interactive learning, goal setting, and daily targets. It offers offline study mode and ad-free studying on its paid plans, making learning a breeze and enabling users to focus on mastering their subjects with ease and confidence.

textgenrnn is a Python 3 module built on Keras/TensorFlow designed for creating character-level recurrent neural networks (char-RNNs). It enables users to easily train text-generating neural networks of any size and complexity on any text dataset. The tool incorporates modern neural network architectures, including attention-weighting and skip-embedding, to accelerate training and enhance model quality. Users can train and generate text at either the character or word level, configure RNN size, layer count, and use bidirectional RNNs. It supports training on generic input text files, including large ones, and allows for GPU-trained models to generate text on a CPU. Additionally, textgenrnn offers a powerful CuDNN implementation for faster GPU training and supports contextual labels for improved learning and results.

The synthetic-personality-dataset offers a high-fidelity collection of 10,000 synthetic records designed to simulate the behavioral and social patterns of introverted and extroverted individuals. Generated using Syncora.ai's synthetic data engine, this dataset ensures zero privacy risk while preserving real-world behavioral distributions. It is ideal for researchers, data scientists, and AI developers focused on personality prediction, behavioral modeling, machine learning experiments, and social science research. The dataset includes features like time spent alone, social event attendance, social media posting habits, and a personality target label, making it suitable for various analytical and ML use cases without compromising privacy or ethical concerns.

TimeCapsuleLLM is an innovative open-source project focused on creating language models (LLMs) trained exclusively on data from specific historical periods and geographic locations. The primary goal is to mitigate modern biases inherent in contemporary LLMs and accurately emulate the linguistic style, vocabulary, and worldview of a chosen era. The project has developed several versions, including v0, v0.5, v1, and v2, with increasing dataset sizes and model parameters, built on architectures like nanoGPT, Phi 1.5, and llamaforcausallm. It emphasizes Selective Temporal Training (STT) where all training data is curated from a defined historical window, ensuring the model's knowledge and language reflect that period without modern influence. The project provides core training scripts, tokenizer building tools, and detailed documentation for researchers and developers interested in historical language modeling.

talk2arxiv is an open-source Retrieval-Augmented Generation (RAG) system specifically designed for academic paper PDFs. It enables users to chat with any ArXiv paper by simply modifying the paper's URL. The system features PDF parsing using GROBID for efficient text extraction, a custom chunking algorithm that organizes text by logical sections and recursive subdivision, and Cohere's EmbedV3 model for accurate text embeddings. It integrates with Qdrant for vector database storage and querying, which also caches research papers to avoid re-embedding. A reranking process ensures contextual relevance based on user input. The frontend is built with Typescript, ReactJS, TailwindCSS, and NextJS, while the backend utilizes Flask, Gunicorn, and Nginx.

tokenizers is an open-source library developed by Hugging Face, offering highly optimized and versatile tokenizers for natural language processing tasks. Implemented primarily in Rust, it boasts exceptional performance, capable of tokenizing a gigabyte of text on a server's CPU in less than 20 seconds. The library supports training new vocabularies and tokenizing text using popular models like Byte-Pair Encoding, WordPiece, and Unigram. It includes features such as alignment tracking during normalization, ensuring that the original sentence segments corresponding to tokens can always be retrieved. Additionally, it handles pre-processing steps like truncation, padding, and adding special tokens required by various models, making it suitable for both research and production environments.

deepmath is a deep-tech company specializing in advanced mathematical modeling, engineering simulations, and AI-enhanced engineering. It provides industry-grade services and next-gen engineering tools by combining advanced mathematical modeling, physics-based simulation, statistics, and AI. The company focuses on solving complex physical and operational challenges in industries such as renewable energy, offshore engineering, and marine engineering, particularly when standard tools or workflows are insufficient. deepmath offers solutions like Finite Element Methods (FEM), Computational Fluid Dynamics (CFD), Discrete Event Simulations (DES), and various forms of Artificial Intelligence (AI) to provide high-fidelity descriptions and predictions for optimizing design and operations. They also offer custom in-house tool development and support for startups and R&D teams.

Trajectory Transformer is an open-source code release that implements offline reinforcement learning as a sequence modeling problem. Based on the paper "Offline Reinforcement Learning as One Big Sequence Modeling Problem," this tool provides a framework for training models to predict trajectories. It includes scripts for training transformers on various datasets and for planning with these models. The project also offers pretrained models for multiple datasets, allowing users to quickly experiment and reproduce results. It supports installation via conda or Docker, and provides utilities for running jobs on Azure, making it suitable for researchers and engineers in reinforcement learning and robotics.

TurboScribe is an AI-powered transcription tool designed to convert audio and video files into text. It leverages advanced AI to provide accurate transcriptions in over 98 languages and offers translation into more than 134 languages. Users can upload files up to 10 hours long or 5 GB in size, with the ability to upload up to 50 files at once for paid users. The platform includes features like bulk exports, all transcription modes, and unlimited storage for paid subscribers. TurboScribe offers a free tier for transcribing up to 3 files daily, each up to 30 minutes, making it accessible for casual users while providing robust features for professionals.

TASO, the Tensor Algebra SuperOptimizer for Deep Learning, significantly enhances the performance of deep neural network models. It achieves this by automatically generating and verifying graph transformations to build a vast search space of computation graphs equivalent to the original DNN model. Employing a cost-based search algorithm, TASO discovers highly optimized computation graphs, leading to up to a 3x performance improvement over graph optimizers in current deep learning frameworks. It supports optimizing pre-trained models in ONNX, TensorFlow, and PyTorch formats, and offers a Python interface for arbitrary DNN architectures. Optimized graphs can be exported to ONNX for use in existing deep learning frameworks, maintaining original model accuracy.

Texar is a comprehensive toolkit designed to support a broad range of machine learning tasks, with a particular focus on natural language processing and text generation. Built on TensorFlow, it offers a rich library of modular and easy-to-use ML components and functionalities, enabling both researchers and practitioners to rapidly prototype and experiment with models. Key features include support for pre-trained models like BERT, GPT2, and XLNet, and full customizability at multiple abstraction levels. Texar is versatile, supporting various tasks, models, algorithms, data processing, and evaluation methods, from encoder-decoder architectures to reinforcement learning and adversarial learning. It emphasizes modularity for maximum re-use and clean APIs, based on a principled decomposition of learning, inference, and model architecture. The toolkit also supports distributed model training with multiple GPUs and provides extensive documentation and examples.

VideoLLaMA2 is an open-source project designed to significantly advance spatial-temporal modeling and audio understanding within video-Large Language Models (LLMs). It offers a comprehensive framework for researchers and developers to explore and build upon state-of-the-art video analysis capabilities. The tool provides various pre-trained models, including vision-only and audio-visual checkpoints, supporting tasks such as multi-choice video QA, video captioning, open-ended video QA, and audio-visual QA. It includes detailed instructions for installation, running online and offline demos, and quick-start guides for training and evaluating custom VideoLLaMA2 models using datasets like VideoLLaVA. The project emphasizes its top performance on leaderboards like MLVU and VideoMME for ~7B-sized VideoLLMs.

VideoLLM-online is the official implementation of an Online Video Large Language Model for Streaming Video, presented at CVPR 2024. Unlike traditional models that process full videos offline, VideoLLM-online enables real-time interaction within a video stream, allowing it to proactively update responses based on activity changes or assist with next steps. It features a cheap and scalable method for synthesizing streaming data by transforming offline annotations into dialogue data using open-source LLMs. The inference method is parallelized, combining video encoding, LLM forwarding, and response generation asynchronously, achieving high speeds of 10-15 FPS on an A100 GPU for long-form videos up to 10 minutes. The tool is designed for researchers and developers working with streaming video analysis and real-time multimodal AI.

Vibe Voice Custom Voices is an innovative audio & music tool hosted on Hugging Face Spaces, designed for generating audio from text input. It offers robust support for both single and multi-speaker voices, making it versatile for various audio production needs. A key feature is its voice cloning capability, allowing users to upload audio clips for each speaker to replicate their voices accurately. The application provides a generated audio output, enabling creators to produce custom voice content efficiently. This tool is ideal for those looking to experiment with voice synthesis and cloning without complex setups, offering an accessible platform for audio creation.

Vietnam Female Voice TTS is a free AI tool hosted on Hugging Face that specializes in converting written Vietnamese text into natural-sounding speech with a female voice. Users can input their desired text directly into the application, and it will generate an audio clip of the text being read aloud. This tool is ideal for a variety of applications, including content creation, educational materials, and accessibility solutions, allowing for easy and quick generation of Vietnamese audio from text. Its straightforward interface makes it accessible for users who need to vocalize Vietnamese content without complex setups.

VideoMamba is an innovative open-source state space model designed for efficient video understanding, specifically addressing the dual challenges of local redundancy and global dependencies in video data. It adapts the Mamba architecture to the video domain, overcoming limitations found in existing 3D convolution neural networks and video transformers. Its linear-complexity operator enables efficient long-term modeling, which is crucial for processing high-resolution and extended video content. The tool demonstrates scalability in the visual domain without requiring extensive dataset pretraining, thanks to a novel self-distillation technique. It also exhibits sensitivity for recognizing fine-grained short-term actions, superiority in long-term video understanding, and compatibility with multi-modal contexts, setting a new benchmark for comprehensive video analysis.

VideoCoF is an AI-powered tool designed for unified video editing, leveraging temporal reasoning to understand and apply changes based on user prompts. Users can upload an input video and specify desired edits through text prompts, and the application will generate a new video incorporating those changes. This capability makes it suitable for various content creation needs, allowing for precise modifications that consider the temporal context of the video. The tool is hosted on Hugging Face Spaces, indicating its accessibility and potential for community-driven development and use.

Thai Sentence Embedding Benchmark is a specialized AI tool designed to evaluate and rank Thai sentence embedding models. It features a comprehensive leaderboard that showcases the performance of different models across a variety of datasets and tasks relevant to the Thai language. Users can access detailed scores for each model, enabling them to compare and select the most suitable embeddings for their specific natural language processing (NLP) applications. This tool is particularly valuable for AI researchers and NLP engineers who require robust benchmarks for developing and optimizing Thai language models.

tts Text To Speech is a powerful text-to-speech (TTS) tool built on Next-gen Kaldi, available as a Hugging Face Space. It allows users to easily convert written text into spoken audio. The application provides options to select from various languages and TTS models, offering flexibility in voice output. Additionally, users can specify a speaker ID and adjust the speaking speed to customize the generated audio. The tool outputs the spoken text as a WAV audio file and also indicates the duration of the generated audio, making it suitable for a range of applications from content creation to research and development.