Experiments for understanding disentanglement in VAE latent representations

Representational Learning of Single-Lead Electrocardiogram Signals using Beta-TCVAE

Applying VAE and DGM families to JATS personality survey database in PyTorch

It's a repo for figuring out ways how to get automatic text summaries to Auto-ML models that perform regression (get compressed and factorized latent representation). Ideally it should be even able to answer questions on properties of that model.

Analyze language model adaptation by tracing semantic hub preservation and token routing during continued pre-training.