AIGoogle Research9 months ago
A collaborative approach to image generation
1 min read
Google Research introduces PASTA, a reinforcement learning agent that refines text-to-image output over multiple interactions by learning user preferences. A novel user simulation technique enables this process. PASTA consistently produced more satisfying images than a baseline model. Google released a dataset of over 7,000 human rater interactions with PASTA.
Level
Hype check
Tap to vote and see what everyone thinks.
