发表机构
CERN（欧洲核子研究中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出粒度邻居组装检索层GNA，通过整窗和逐变量双粒度检索及门控融合，在96个设置中85个提升Transformer预测性能，优于无检索基线。
AI 中文摘要
深度预测器从固定长度的回看窗口进行预测，延长该窗口会带来收益递减且成本不断增加。检索增强则通过展示相似的过去情境如何延续来帮助模型。检索整个过去窗口会使每个变量都获得同一过去时刻的延续。然而，在多变量序列中，最佳的历史匹配因变量而异。我们提出了GNA（粒度邻居组装），一种用于预测骨干网络的检索层，它在两个粒度上组装邻居：整个过去窗口（保持变量间的连贯性）和每个变量各自的邻居（每个变量从自身最佳匹配的历史中获取其未来）。一个学习得到的门控机制在每个预测步骤和每个变量上决定，相对于持久性预测以及骨干网络自身的预测，对这些未来值的信任程度。候选值来自一个经过训练的嵌入，该嵌入用于预测每个窗口的未来，且检索严格因果：一个过去窗口仅在其未来被观测到后才被使用。在保持每个模型相同的回看长度和所有数据集相同的检索常数的条件下，GNA在96个数据集-预测水平设置中的85个上改进了两个Transformer骨干网络，在12个标准基准中的8个上取得了最低MSE，并在其中10个基准的每个随机种子下均优于其骨干网络。两种粒度都是必需的，且与查询不匹配的邻居比没有邻居更差。检索在回看信息最少的地方帮助最大：门控将信任转向更远未来的检索结果。在失败的情况下，即长预测水平上的小时级非平稳序列，损失与检索未来值的漂移水平一致。
英文摘要
Deep forecasters predict from a fixed-length lookback window, and lengthening it gives diminishing returns at a growing cost. Retrieval augmentation instead shows the model how similar past situations continued. Retrieving a whole past window gives every variate the continuation of the same past moment. In multivariate series, however, the best past match differs from variate to variate. We present GNA (Granular Neighbor Assembly), a retrieval layer for forecasting backbones that assembles neighbors at two granularities: whole past windows, which keep the variates coherent, and per-variate neighbors, in which each variate takes its future from its own best-matching past. A learned gate decides, per forecast step and variate, how much to trust these futures against a persistence forecast, next to the backbone's own forecast. Candidates come from an embedding trained to predict each window's future, and retrieval is strictly causal: a past window is used only once its future has been observed. With the same lookback for every model and the same retrieval constants for all datasets, GNA improves two Transformer backbones in 85 of 96 dataset-horizon settings, gives the lowest MSE on 8 of 12 standard benchmarks and beats its backbone in every seed on 10 of them. Both granularities are needed, and neighbors of mismatched queries are worse than none. Retrieval helps most where the lookback says least: the gate shifts trust to retrieved futures further ahead. Where it fails, on hourly non-stationary series at long horizons, the loss is consistent with a drifting level of the retrieved futures.
