Future deep learning models will be distinguished by systems that perpetually learn through interaction, imagination, and cooperation, blurring the line between training and inference. This makes continual learning a critical challenge, as methods that efficiently maximize bidirectional transfer across learning trajectories will be essential. Replay is on track to play a foundational role in continual learning, allowing models to directly reconcile new information with past knowledge. In practice, however, replay is quite unscalable, doubling the cost of continual learning when applied naively. Moreover, the continual learning literature has not fully synchronized with the multi-task fine-tuning literature, having not fully integrated highly scalable techniques like model merging and low rank adaptation into a replay-enabled toolset that can produce a unified model in the face of many sequential tasks. In this paper, we begin by applying and analyzing low rank adaptation in a continual learning setting. Next, we introduce consolidation, a phasic approach to replay which leads to up to 55\% less replay samples being needed for a given performance target. Then, we propose sequential merging, an offshoot of task arithmetic which is tailored to the continual learning setting and is shown to work well in combination with replay. Finally, we demonstrate that the developed strategies can operate synergistically, resulting in a highly scalable toolset that outperforms standalone variants.

The web forum is a key tool in the building of new knowledge among students in Learning Management Systems. Students’ posted messages, in fact, build up a relationship network which supports a collaborative reflection about the forum topic. In this network two interaction levels can be distinguished. The former is the interaction between peers (the students), the latter between students and instructors (teachers and tutors). The role of the second interaction is particularly important as a feedback mechanism in the discussion dynamic but it is subjected to two kinds of limitations. The first one is the huge number of messages that makes difficult, for tutors and teachers, to quickly evaluate the progress of their students and the second one is the subjective bias of the tutors that influence the evaluation. In order to limit these two inefficiencies a multiagent system can be used to monitor such evolution and recognize the state in which the forum is. Such system is based on metrics derived from the textual and social network analysis that, feeding a rule engine, gives the instructor a more objective view of the forum evolution.