Gido M. van de Ven

Co-observation in training data can dramatically enhance generalization in continual learning, revealing a new dimension beyond forgetting and plasticity.

Forget catastrophic forgetting: modular memory, blending in-context and in-weight learning, offers a practical path to truly continual learning agents.

Weight regularization, often overlooked in parameter-efficient continual learning, can still significantly improve the stability-plasticity trade-off, even when using low-rank adapters.