In always-on HAR deployments, model accuracy erodes silently as domain shift accumulates over time. Addressing this challenge requires moving beyond one-off updates toward instance-driven adaptation from streaming data. However, continuous adaptation exposes a fundamental tension: systems must selectively learn from informative instances while actively forgetting obsolete ones under long-term, non-stationary drift. To address them, we propose CODA, a continuous online adaptation framework for mobile sensing. CODA introduces two synergistic components: (i) Cache-based Selective Assimilation, which prioritizes informative instances likely to enhance system performance under sparse supervision, and (ii) an Adaptive Temporal Retention Strategy, which enables the system to gradually forget obsolete instances as sensing conditions evolve. By treating adaptation as a principled cache evolution rather than parameter-heavy retraining, CODA maintains high accuracy without model reconfiguration. We conduct extensive evaluations on four heterogeneous datasets spanning phone, watch, and multi-sensor configurations. Results demonstrate that CODA consistently outperforms one-off adaptation under non-stationary drift, remains robust against imperfect feedback, and incurs negligible on-device latency.

翻译：摘要：在始终开启的HAR部署中，随着领域偏移随时间累积，模型精度会悄然下降。解决这一挑战需要从一次性更新转向基于流式数据的实例驱动自适应。然而，持续自适应暴露了一个根本性矛盾：系统必须在长期非平稳漂移下，有选择地从信息性实例中学习，同时主动遗忘过时实例。为此，我们提出CODA——一种面向移动传感的持续在线自适应框架。CODA引入两个协同组件：(i) 基于缓存的选择性同化机制，优先选择在稀疏监督下可能提升系统性能的信息性实例；(ii) 自适应时序保留策略，使系统能在传感环境演进中逐步遗忘过时实例。通过将自适应视为原则性缓存演化而非参数密集型重训练，CODA无需模型重构即可保持高精度。我们在涵盖手机、手表及多传感器配置的四个异构数据集上开展广泛评估。结果表明，CODA在非平稳漂移下持续优于一次性自适应方法，对不完美反馈保持稳健，且引入的终端时延可忽略不计。