Spectral Transformer Neural Processes

📰 ArXiv cs.AI

arXiv:2605.09498v1 Announce Type: cross Abstract: Time series, spatial data, and images are natural applications of Neural Processes. However, when such data exhibit strong periodicity and quasi-periodicity, existing methods often suffer from underfitting and generalise poorly beyond the training distribution. In this work, we propose Spectral Transformer Neural Processes (STNPs), a frequency-aware extension of Transformer Neural Processes (TNPs). STNPs introduce a Spectral Aggregator that estim

Published 12 May 2026

Full Article

Title: Spectral Transformer Neural Processes

Abstract:
arXiv:2605.09498v1 Announce Type: cross Abstract: Time series, spatial data, and images are natural applications of Neural Processes. However, when such data exhibit strong periodicity and quasi-periodicity, existing methods often suffer from underfitting and generalise poorly beyond the training distribution. In this work, we propose Spectral Transformer Neural Processes (STNPs), a frequency-aware extension of Transformer Neural Processes (TNPs). STNPs introduce a Spectral Aggregator that estim
Read full paper → ← Back to Reads