Assess lightweight adaptation generalizability

Background: Forecasting models developed and validated in one urban setting, like Beijing, may not perform optimally when applied to other cities due to differences in emission profiles, atmospheric conditions, or urban structures.

Question / Future Work: Extend the current adaptation strategies (walk-forward refitting and online residual correction) to heterogeneous datasets, different cities, and varying sensing conditions to assess the broader applicability and generalizability of the lightweight forecasting approaches tested.

Why It Matters: Determining the generalizability of lightweight forecasting methods is crucial for their adoption in diverse global urban air quality management systems.

Evidence: Extending lightweight adaptation strategies to heterogeneous datasets, cities, and sensing conditions will be important for assessing the broader applicability of the proposed approach.