Genie: Generative Interactive Environments (Paper Walkthrough) скачать в хорошем качестве

Genie: Generative Interactive Environments (Paper Walkthrough)

📖Paper: https://arxiv.org/abs/2402.15391 🐸RibbitRibbit: https://ribbitribbit.co/paper/arxiv.2... 🌍Genie3: https://deepmind.google/discover/blog... 👥Authors: Jake Bruce, Michael Dennis, Ashley Edwards, Jack Parker-Holder, Yuge Shi, Edward Hughes, Matthew Lai, Aditi Mavalankar, Richie Steigerwald, Chris Apps, Yusuf Aytar, Sarah Bechtle, Feryal Behbahani, Stephanie Chan, Nicolas Heess, Lucy Gonzalez, Simon Osindero, Sherjil Ozair, Scott Reed, Jingwei Zhang, Konrad Zolna, Jeff Clune, Nando de Freitas, Satinder Singh, Tim Rocktäschel 🏫Institutes: Google DeepMind; University of British Columbia Genie Turns Prompts into Interactive Worlds Using Latent Actions Google just released Genie 3 — so we're revisiting the original Genie paper from 2024. Researchers at Google DeepMind and UBC trained a model to turn sketches, photos, or text prompts into interactive, playable video worlds — using only unlabeled internet videos. No action labels. Just 8 latent codes to control everything. Want to discover more AI papers like this? 🚀 Head over to https://RibbitRibbit.co 🐸 — Discover Research The Fun Way!