Projected Latent Distillation for Data-Agnostic Consolidation in distributed continual learning | Publicación