转型中的持续学习

摘要

经典的持续学习（CL）主要聚焦于通过以参数为中心的机制来更新和保持知识，例如训练策略、架构设计和权重自适应。然而，新兴范式正在重塑持续学习的范畴，使其超越传统的模型适应性视角。例如，在策略学习拓展了更新机制的维度；测试时训练将持续学习从训练阶段延伸至推理阶段；而外部辅助组件，如记忆、技能库和交互协议，将模型能力的演化边界远远拓展至静态参数空间之外。总体而言，这些进展表明学习正从以参数为中心向系统级适应性转变。为刻画这一转变，我们通过三个维度审视持续学习的演化过程：学习发生的“何时”（When）、“如何”（How）与“何处”（Where）。其中，“如何”维度涵盖离策略、在策略及超越梯度的优化机制；“何时”维度涵盖预训练、后训练及推理时阶段的演化；“何处”维度则界定内部参数更新与外部结构约束之间的区别。以这一三轴框架为基础，我们对代表性方法进行了系统梳理，追踪持续学习正在经历的转变，并讨论了这一范式转移所带来的关键挑战、更广泛的影响以及未来发展方向。

English

Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through parameter-centric mechanisms, e.g., training strategies, architectural designs, and weight adaptation. However, emerging paradigms are reshaping the scope of CL beyond this traditional model adaptation view. For instance, on-policy learning broadens the space of update mechanisms; test-time training extends CL from the training phase to inference; and external harness components such as memory, skill libraries, and interaction protocols extend the evolutionary boundaries of model capabilities far beyond the static parameter space. Collectively, these developments indicate a transition from parameter-centric learning toward system-level adaptation. To characterize this transition, we examine the evolution of continual learning through three dimensions: When, How, and Where learning occurs. The How dimension encompasses off-policy, on-policy, and beyond-gradient optimization mechanics. The When dimension captures evolution across pre-training, post-training, and inference-time stages. The Where dimension delineates updates occurring within internal parameters versus external structural constraints. Anchored by this tri-axial framework, we systematically survey representative methods, trace the ongoing transition of continual learning, and discuss the key challenges, broader implications, and future directions arising from this paradigm shift.