swiftyos / auto-gpt-benchmarks Goto Github PK

View Code? Open in Web Editor NEW

A repo built for the purpose of benchmarking the performance of agents, regardless of how they are set up and how they work.

License: MIT License

Shell 0.07% Python 46.36% HTML 0.13% Jupyter Notebook 53.43%

auto-gpt-benchmarks's Introduction

About Me

I'm an AI Engineer with a background in Aerospace Engineering, specializing in the development of intelligent agents. My expertise in Python and Rust allows me to build AI systems that are not only powerful but also efficient and scalable.

Projects

AutoGPT

GitHub Repo: AutoGPT
Contributed significantly to the AutoGPT project. AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

Published Papers

GAIA: a benchmark for General AI Assistants
Authors: Grégoire Mialon, Clémentine Fourrier, Craig Swift, Thomas Wolf, Yann LeCun, Thomas Scialom
arXiv Link: 2311.12983
Introduced GAIA, a benchmark for evaluating general AI assistants, focusing on complex challenges like reasoning and multi-modality, presented at ICLR.
Testing Language Model Agents Safely in the Wild
Authors: Silen Naihin, David Atkinson, Marc Green, Merwane Hamadi, Craig Swift, Douglas Schonholtz, Adam Tauman Kalai, David Bau
arXiv Link: 2311.10538
Proposed a framework for conducting safe tests of autonomous agents in real-world settings, highlighting the integration of safety and reliability, presented at NeurIPS.