重新思考持续学习中的迁移：基于重放的实现

中文摘要 AI 辅助

持续学习研究已部署的语言模型如何在无需从头昂贵重训的情况下持续获取新任务。现有方法，无论是基于重放（回放存储的过去数据）还是无重放（正则化或隔离参数），大多只针对一个目标：防止灾难性遗忘。正向迁移，即过去帮助未来，几乎仅通过参数重用实现，且完全未明确何时应预期迁移。我们提前一步：在设计迁移机制前，先问何时应存在迁移。我们用三个可衡量条件的框架作答：目标任务在其自身有限监督之外必须有改进空间，可迁移信息在持续优化中必须留存，重放必须来自兼容的先前任务。我们将此观点实例化为迁移选择性重放（TSR），它选择预计能使新任务受益的重放数据而非随意重放过去示例。选择由零训练任务签名引导，而蒸馏保持先前任务的稳定性。在低预算模式下的标准持续学习协议中，TSR持续改进正向迁移同时保持稳定性，在异构和同构任务流中均优于现有重放基线。更广泛地说，结果表明应将迁移视为持续学习的首要目标，在设计前先理解它。

英文摘要

Continual learning studies how deployed language models can continually acquire new tasks without expensive retraining from scratch. Existing methods, whether rehearsal-based (replaying stored past data) or rehearsal-free (regularising or isolating parameters), overwhelmingly target one objective: preventing catastrophic forgetting. Forward transfer, the past helping the future, has meanwhile been pursued almost exclusively through parameter reuse, with no explicit account of when transfer should be expected at all. We begin one step earlier: before designing a transfer mechanism, we ask when transfer should exist at all. We answer with a framework of three measurable conditions: the target task must leave room for improvement beyond its own limited supervision, transferable information must survive continued optimisation, and replay must come from compatible previous tasks. We instantiate this view as Transfer-Selective Replay (TSR), which selects replay data predicted to benefit the incoming task rather than replaying past examples indiscriminately. Selection is guided by a zero-training task signature, while distillation preserves stability on previous tasks. Under the standard continual learning protocol in the low-budget regime, TSR consistently improves forward transfer while maintaining stability, outperforming existing replay baselines across heterogeneous and homogeneous task streams. More broadly, the results argue for treating transfer as a first-class objective of continual learning, to be understood before it is engineered.

发表机构

• University of Chicago（芝加哥大学）

机构由 AI 辅助整理，请以论文原文为准。