-
, feature matching in self-supervised audio representations, and reference-free audio quality prediction. Integrating watermarking capabilities into deep audio generative models, including extensions
-
work by further developing the code to include time dependent cohesive bonding between the ice blocks. Then you will use the code to model ice pile-up. The work will have a multi-scale component as you
-
, deepfake detection, and generative audio. A strong open-science environment: the project aims to publish in leading venues and release software source code and trained models to support reproducibility and
-
or similar. Running social simulations with AI agents using platforms like Concordia or similar. Coding in Python or a similar language for simulation, visualization, and advanced statistical analysis (causal