AI 中文总结
KReF是一种无训练检索框架，将历史未来作为查询的局部经验预测分布，在多数LTSF基准上取得了优于或匹配训练基线的预测性能，确立了检索作为LTSF的有效归纳偏置的价值。
AI 中文摘要
概率长期时间序列预测通常依赖于经过训练的模型。无训练的保形方法通常在预先存在的点预测器周围构建区间，且无法天然表示完整的预测分布；其序列变体在长预测区间下还会面临反馈延迟不断增加的问题。我们提出了KReF，一种无训练检索框架，该框架将检索到的历史未来视为查询的局部经验预测分布。经过稳健预处理后，KReF使用手工统计特征或冻结的随机傅里叶特征嵌入每个回溯窗口，并检索相似的历史回溯-未来对。这些对的相似度权重直接定义了预测质量、分位数、连续排名概率得分（CRPS）以及加权平均点预测。KReF还利用观测到的查询回溯窗口构建概率积分变换图，并应用经验证选择的扩展和收缩率来调整区间边界。在六个长期时间序列预测（LTSF）基准数据集和四个预测区间下，KReF在全部12种数据集-嵌入设置中取得了最低的CRPS，在9种设置中取得了最低的90%区间得分（IS90）。在六个数据集中的两个上，其点预测也与训练过的基线相当或更优。存档- oracle分析进一步显示，在更精细的按预测区间和通道路由设置下，仍有巨大提升空间。这些结果确立了检索作为LTSF的一种有用且未被充分探索的归纳偏置的价值。
英文摘要
Probabilistic long-term time-series forecasting commonly relies on trained models. Training-free conformal methods typically construct intervals around a pre-existing point forecaster and do not natively represent a complete predictive distribution; sequential variants additionally suffer from increasingly delayed feedback at long horizons. We propose KReF, a training-free retrieval framework that treats retrieved historical futures as a querylocal empirical predictive distribution. After robust preprocessing, KReF embeds each lookback using handcrafted statistics or frozen random Fourier features and retrieves similar historical lookback-future pairs. Their similarity weights directly define predictive masses, quantiles, CRPS, and a weighted-mean point forecast. KReF further uses the observed query lookback to construct a probability-integral-transform map and applies validation-selected expansion and shrinkage rates to adapt interval boundaries. Across six LTSF benchmarks and four horizons, KReF obtains the lowest CRPS in all 12 dataset-embedding settings and the lowest IS90 in 9 settings. Without gradient-based fitting, its point forecasts also match or surpass trained baselines on two of six datasets. An archive-oracle analysis further reveals substantial headroom under finer horizon- and channel-wise routing. These results establish retrieval as a useful and underexplored inductive bias for LTSF.
