AI & ML
Video generation models as world simulators
OpenAI Blog0 views
We explore large-scale training of generative models on video data. Specifically, we train text-conditional diffusion models jointly on videos and images of variable durations, resolutions and aspect ratios. We leverage a transformer architecture that operates on spacetime patches of video and image latent codes. Our largest model, Sora, is capable of generating a minute of high fidelity video. Our results suggest that scaling video generation models is a promising path towards building general purpose simulators of the physical world.
Source:https://openai.com/index/video-generation-models-as-world-simulatorsRelated posts
I'm a Second-Year CS Student Building Windows Software. Am I on the Right Track?
0dev.to
Code Review Anxiety: 6 Cognitive Distortions That Hit During PR Review (And How to Break Them)
0dev.to
When internship applications open by industry
0dev.to
LLM agents lose confidence under noisy retrieval
0dev.to
Tree-Ising Beats the GPT-6 Astra Weight Edit for $0.90
0dev.to
A tiny charge can be a test of your card
0dev.to