Are Time Series Foundation Models Susceptible to Catastrophic Forgetting?
This addresses a robustness issue for users of TSFMs in continual learning scenarios, though it is incremental as it explores an underexplored aspect of existing models.
The study investigated whether Time Series Foundation Models (TSFMs) experience catastrophic forgetting when fine-tuned sequentially on multiple datasets, finding that fine-tuning improves new task performance but often causes significant degradation on prior tasks, highlighting a stability-plasticity dilemma.
Time Series Foundation Models (TSFMs) have shown promising zero-shot generalization across diverse forecasting tasks. However, their robustness to continual adaptation remains underexplored. In this work, we investigate the extent to which TSFMs suffer from catastrophic forgetting when fine-tuned sequentially on multiple datasets. Using synthetic datasets designed with varying degrees of periodic structure, we measure the trade-off between adaptation to new data and retention of prior knowledge. Our experiments reveal that, while fine-tuning improves performance on new tasks, it often causes significant degradation on previously learned ones, illustrating a fundamental stability-plasticity dilemma.