TS-Memory: Plug-and-Play Memory for Time Series Foundation Models
This addresses the problem of distribution shift adaptation for users of time series models, offering a plug-and-play solution that balances performance and efficiency, though it is incremental as it builds on existing parametric and non-parametric methods.
The paper tackles the challenge of adapting Time Series Foundation Models to downstream domains under distribution shift by proposing TS-Memory, a lightweight memory adapter that improves point and probabilistic forecasting with efficiency comparable to frozen backbones.
Time Series Foundation Models (TSFMs) achieve strong zero-shot forecasting through large-scale pre-training, but adapting them to downstream domains under distribution shift remains challenging. Existing solutions face a trade-off: Parametric Adaptation can cause catastrophic forgetting and requires costly multi-domain maintenance, while Non-Parametric Retrieval improves forecasts but incurs high inference latency due to datastore search. We propose Parametric Memory Distillation and implement it as TS-Memory, a lightweight memory adapter that augments frozen TSFMs. TS-Memory is trained in two stages. First, we construct an offline, leakage-safe kNN teacher that synthesizes confidence-aware quantile targets from retrieved futures. Second, we distill this retrieval-induced distributional correction into a lightweight memory adapter via confidence-gated supervision. During inference, TS-Memory fuses memory and backbone predictions with constant-time overhead, enabling retrieval-free deployment. Experiments across diverse TSFMs and benchmarks demonstrate consistent improvements in both point and probabilistic forecasting over representative adaptation methods, with efficiency comparable to the frozen backbone.