跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.26355 · v1

Mamba Flow Matching Neural Processes: Linear-Time Inference for Irregularly Observed Spatial Fields

首次提交 2026-11-20 00:25
版本更新 2026-11-20 00:25
雷达观测 2026-11-20 12:52
arXiv ↗PDF ↗

摘要

Interpolating a spatial field, such as daily maximum temperature, from sparse, irregularly located observations recurs across the environmental and health sciences. Standard Gaussian-process kriging refits hyperparameters and solves an O(N3)O(N^3) system for every new field. When many related fields are observed, such as daily maps from one sensor network, neural processes (NPs) instead learn the interpolation map from past fields and predict a new field in one forward pass. Joint-predictive variants, which sample whole fields rather than independent per-location values, typically use self-attention over context and query points, costing O(K2)O(K^2) in the combined point count KK and limiting practical use to a few thousand points. State-space models offer a linear-time alternative, but a direct state-space adaptation compresses the context into one hidden state before reading queries, reducing what the decoder sees. We propose MambaFlowNP, a linear-compute joint-predictive NP trained with conditional flow matching. A multi-directional Mamba-2 scan reads context and query tokens in one shared sequence, so each query's velocity is read from its own scan position rather than a compressed context summary, and because queries only read the scan state, predictions are consistent across query sets. MambaFlowNP is competitive with the strongest attention-based joint-predictive baselines on synthetic Matérn Gaussian-process benchmarks, and on regional to continental-scale NOAA temperature interpolation it attains the lowest RMSE and CRPS and near-nominal coverage on every tier, while running 114--784×\times faster than the strongest joint-predictive baseline. A single forward pass scales to 10610^6 points on one NVIDIA L40S GPU in under a second. This makes joint prediction feasible for continental-scale, irregularly sampled sensor networks; predictive quality is evaluated up to K=5,580K=5{,}580.

作者备注:Accepted at NeurIPS 2026. XX pages, X figures. Code: https://github.com/Cosmonaught/Mambaflownp

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-20 00:25 · 最新观测版本

Mamba Flow Matching Neural Processes: Linear-Time Inference for Irregularly Observed Spatial Fields

Interpolating a spatial field, such as daily maximum temperature, from sparse, irregularly located observations recurs across the environmental and health sciences. Standard Gaussian-process kriging refits hyperparameters and solves an O(N3)O(N^3) system for every new field. When many related fields are observed, such as daily maps from one sensor network, neural processes (NPs) instead learn the interpolation map from past fields and predict a new field in one forward pass. Joint-predictive variants, which sample whole fields rather than independent per-location values, typically use self-attention over context and query points, costing O(K2)O(K^2) in the combined point count KK and limiting practical use to a few thousand points. State-space models offer a linear-time alternative, but a direct state-space adaptation compresses the context into one hidden state before reading queries, reducing what the decoder sees. We propose MambaFlowNP, a linear-compute joint-predictive NP trained with conditional flow matching. A multi-directional Mamba-2 scan reads context and query tokens in one shared sequence, so each query's velocity is read from its own scan position rather than a compressed context summary, and because queries only read the scan state, predictions are consistent across query sets. MambaFlowNP is competitive with the strongest attention-based joint-predictive baselines on synthetic Matérn Gaussian-process benchmarks, and on regional to continental-scale NOAA temperature interpolation it attains the lowest RMSE and CRPS and near-nominal coverage on every tier, while running 114--784×\times faster than the strongest joint-predictive baseline. A single forward pass scales to 10610^6 points on one NVIDIA L40S GPU in under a second. This makes joint prediction feasible for continental-scale, irregularly sampled sensor networks; predictive quality is evaluated up to K=5,580K=5{,}580.

查看此版本解读 →查看此版本原文 ↗