Merging on the Fly Without Retraining: ASequential Approach to Scalable Continual Model Merging

Open in new window