Discrete Prototypical Memories for Federated Time Series Foundation Models

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Deng, Liwei, Liu, Qingxiang, Niu, Xinhe, Chen, Shengchao, Sun, Sheng, Wu, Yuankai, Long, Guodong, Liang, Yuxuan
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914448139091968
author Deng, Liwei
Liu, Qingxiang
Niu, Xinhe
Chen, Shengchao
Sun, Sheng
Wu, Yuankai
Long, Guodong
Liang, Yuxuan
author_facet Deng, Liwei
Liu, Qingxiang
Niu, Xinhe
Chen, Shengchao
Sun, Sheng
Wu, Yuankai
Long, Guodong
Liang, Yuxuan
contents Leveraging Large Language Models (LLMs) as federated learning (FL)-based time series foundation models offers a promising way to transfer the generalization capabilities of LLMs to time series data while preserving access to private data. However, the semantic misalignment between time-series data and the text-centric latent space of existing LLMs often leads to degraded performance. Meanwhile, the parameter-sharing mechanism in existing FL methods model heterogeneous cross-domain time-series data into a unified continuous latent space, which contradicts the fact that time-series semantics frequently manifest as discrete and recurring regimes. To address these limitations, we propose \textsc{FeDPM}, a federated framework for time-series foundation models based on discrete prototypical memories. Specifically, we learn local prototypical memory priors for intra-domain time-series data. We then align cross-domain memories to promote a unified discrete latent space and introduce a domain-specific memory update mechanism to balance shared and personalized prototypical knowledge. Extensive experiments demonstrate the efficiency and effectiveness of \textsc{FeDPM}. The code is publicly available at https://anonymous.4open.science/r/FedUnit-64D1.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04475
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Discrete Prototypical Memories for Federated Time Series Foundation Models
Deng, Liwei
Liu, Qingxiang
Niu, Xinhe
Chen, Shengchao
Sun, Sheng
Wu, Yuankai
Long, Guodong
Liang, Yuxuan
Machine Learning
Artificial Intelligence
Leveraging Large Language Models (LLMs) as federated learning (FL)-based time series foundation models offers a promising way to transfer the generalization capabilities of LLMs to time series data while preserving access to private data. However, the semantic misalignment between time-series data and the text-centric latent space of existing LLMs often leads to degraded performance. Meanwhile, the parameter-sharing mechanism in existing FL methods model heterogeneous cross-domain time-series data into a unified continuous latent space, which contradicts the fact that time-series semantics frequently manifest as discrete and recurring regimes. To address these limitations, we propose \textsc{FeDPM}, a federated framework for time-series foundation models based on discrete prototypical memories. Specifically, we learn local prototypical memory priors for intra-domain time-series data. We then align cross-domain memories to promote a unified discrete latent space and introduce a domain-specific memory update mechanism to balance shared and personalized prototypical knowledge. Extensive experiments demonstrate the efficiency and effectiveness of \textsc{FeDPM}. The code is publicly available at https://anonymous.4open.science/r/FedUnit-64D1.
title Discrete Prototypical Memories for Federated Time Series Foundation Models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2604.04475