Search Results for "human"

...Built from the ground up for scalability, composability, and research flexibility, fairseq2 supports a broad range of language, speech, and multimodal content generation tasks, including instruction fine-tuning, reinforcement learning from human feedback (RLHF), and large-scale multilingual modeling. Unlike the original fairseq—which evolved into a large, monolithic codebase—fairseq2 introduces a clean, plugin-oriented architecture designed for long-term maintainability and rapid experimentation. It supports multi-GPU and multi-node distributed training using DDP, FSDP, and tensor parallelism, capable of scaling up to 70B+ parameter models. ...

...The research prototype includes components for visual grounding (understanding when a user references something in view), gesture recognition and synthesis, and turn-taking mechanisms that mirror human conversational timing. Because latency and synchronization are critical, the codebase invests in asynchronous scheduling, overlap of perception and reasoning, and fast fallback responses.

TEN (Transformative Extensions Network) is an open source framework designed to empower developers to build real-time multimodal AI agents capable of voice, video, text, image, and data-stream interaction with ultra-low latency. It includes a full ecosystem, TEN Turn Detection, TEN Agent, and TMAN Designer, allowing developers to rapidly assemble human-like, responsive agents that can see, speak, hear, and interact. With support for languages like Python, C++, and Go, it offers flexible deployment on both edge and cloud environments. Using components like graph-based workflow design, drag-and-drop UI (via TMAN Designer), and reusable extensions such as real-time avatars, RAG (Retrieval-Augmented Generation), and image generation, TEN enables highly customizable, scalable agent development with minimal code.

AI File Sorter is a cross-platform desktop application that uses AI (local LLMs run on your computer) to organize files and suggest meaningful file names based on real content, not just filenames or extensions. The app can analyze images locally and propose descriptive rename suggestions (for example, IMG_2048.jpg → clouds_over_lake.jpg). It can also analyze document text to improve categorization and renaming. Supported formats include PDF, DOCX, XLSX, PPTX, ODT, ODS, ODP, and common...

...KM is not intended to replace human experts, it is built to multiply abilities. Machine should not be responsible for decisions. The IMK is designed to create ready-to-use software applications using simple text files. Any human knowledge can be uploaded to KM by expert not familiar with software coding. Demos present in kit. Good luck!

...The system couples fast GPU policy inference with CPU or GPU-assisted tree search so priors from the network guide exploration while search refines local tactics. Training pipelines mix supervised learning from human professional games and self-play fine-tuning, allowing the model to learn opening patterns and endgame tactics beyond simple pattern libraries. The codebase includes tools for parsing classic Go formats, generating training examples, and evaluating models on standard test suites and online servers. A KGS/online client and match runner make it practical to stage controlled tournaments or continuous rating evaluation. ...

[ Please read https://sourceforge.net/p/darkbot/news/2014/01/darkbots-revitalization/ ] Darkbot is a portable IRC chat robot written in the C language that can be taught responses to user inquiries, and even have conversations with them. Darkbot was originally created by Jason Hamilton as an aid for help channels on Intenet Relay Chat.

To design a complete, wheeled robot, capable of accomplishing simple tasks, manipulate objects and have human interaction capabilities. Includes electronic board design, firmware and AI programming, computer-robot interaction software and mechanics.

ExPo: Exchange Portal is an online financial trading exchange platform designed to run controlled scientific trading experiments between human traders and automated trader robots.

Real time face tracking and recognition refers to the task of locating human faces in a video stream and identifying the faces by matching them against the database of known faces.

MERP is a modular low cost robot platform designed as a research tool or for personal use. It allows real world testing of algorithms developed in simulation for trajectory planning, obstacle avoidance, human robot interaction, among others.

Yacca is an advanced chat bot (other plugins are under development, including a command-line only version for you to interact with yacca) that is designed to resemble a human chatter as closely as possible.

The development of a biometric system and applications e.g. Access Control based on the system to recognize and verify people. The primary goal is face recognition, but other human attributes might get used too in the future.

A software package for evolving neural networks on several e-puck robots (www.e-puck.org, www.glowbots.com). The intention is to let the e-pucks evolve communication in cooperation with human beings.

OpenIA is a proyect whit the thinking machine research objetive. The prexistings human brain knoledges are a base to it. OpenIA es un proyecto que busca conseguir una máquina pensante basandose en los conocimientos preexistentes sobre la mente humana