Tag
This paper introduces Atelier, a method that plans explicit control states before generation to prevent text-to-image models from falling back on artist-name shortcuts, and presents ArtIntentBench for evaluating artist-grounded style control.
OpenVoice is an open-source instant voice cloning model with style and language control, now available on GitHub.
This paper presents a framework for training reinforcement learning agents that can be coached in real-time to adopt different styles while performing core tasks, demonstrated in Horizon Forbidden West, Gran Turismo, and a humanoid walking domain.
Krea 2 is an image model released for enhanced style control and moodboard creation.