让时间序列模型快速适应新数据，不改主干、不慢推理。
TS-Memory: Plug-and-Play Memory for Time Series Foundation Models
- 用知识蒸馏把检索结果转成轻量记忆模块，替代复杂搜索。
- 在多个数据集上点预测和概率预测均优于现有方法，且推理速度不变。
- 适合需要快速适配新领域的工业时间序列场景。
时间序列基础模型（TSFMs）通过大规模预训练实现强大的零样本预测，但在分布偏移下的下游领域适应仍具挑战。现有方法存在权衡：参数化适配易导致灾难性遗忘，需维护多领域参数；非参数检索虽提升预测精度，但推理延迟高。本文提出参数化记忆蒸馏，并实现为轻量级记忆适配器 TS-Memory，可无缝接入冻结的 TSFMs。TS-Memory 分两阶段训练：首先构建离线、无泄露的 kNN 教师，从检索到的未来序列合成置信度感知的分位数目标；其次通过置信度门控监督，将检索引发的分布修正蒸馏至轻量记忆模块。推理时，TS-Memory 以常数时间开销融合记忆与主干预测，实现无需检索部署。在多种 TSFMs 与基准测试上，相较代表性适配方法，点预测与概率预测均取得一致提升，效率与冻结主干相当。
原文摘要 · Abstract (English)
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, retrieval-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. Code: https://github.com/sisuolv/TS-Memory.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。