Controlling Language and Diffusion Models by Transporting Activations
Machine Learning Research at Apple
APRIL 9, 2025
Large generative models are becoming increasingly capable and more widely deployed to power production applications, but getting these models to produce exactly what's desired can still be challenging. Fine-grained control over these models' outputs is important to meet user expectations and to mitigate potential misuses, ensuring the models' reliability and safety.
Let's personalize your content