Modular Memory is the Key to Continual Learning Agents
Abstract
Foundation models have transformed machine learning through large-scale pretraining and increased test-time compute. Despite surpassing human performance in several domains, these models remain fundamentally limited in continuous operation, experience accumulation, and personalization, capabilities that are central to adaptive intelligence. While continual learning research has long targeted these goals, its historical focus on in-weight learning (IWL), i.e., updating a single model's parameters to absorb new knowledge, has rendered catastrophic forgetting a persistent challenge. Our position is that combining the strengths of In-Weight Learning (IWL) and the newly emerged capabilities of In-Context Learning (ICL) through the design of modular memory is the missing piece for continual adaptation at scale. We outline a conceptual framework for modular memory-centric architectures that leverage ICL for rapid adaptation and knowledge accumulation, and IWL for stable updates to model capabilities, charting a practical roadmap toward continually learning agents.
Cite
@article{arxiv.2603.01761,
title = {Modular Memory is the Key to Continual Learning Agents},
author = {Vaggelis Dorovatas and Malte Schwerin and Andrew D. Bagdanov and Lucas Caccia and Antonio Carta and Laurent Charlin and Barbara Hammer and Tyler L. Hayes and Timm Hess and Christopher Kanan and Dhireesha Kudithipudi and Xialei Liu and Vincenzo Lomonaco and Jorge Mendez-Mendez and Darshan Patil and Ameya Prabhu and Elisa Ricci and Tinne Tuytelaars and Gido M. van de Ven and Liyuan Wang and Joost van de Weijer and Jonghyun Choi and Martin Mundt and Rahaf Aljundi},
journal= {arXiv preprint arXiv:2603.01761},
year = {2026}
}
Comments
This work stems from discussions held at the Dagstuhl seminar on Continual Learning in the Era of Foundation Models (October 2025)