Enhancing few-shot time series forecasting with LLM-guided diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Haonan, Shuai, Dehua, Wang, Liming, Liu, Xiyang, Tian, Long
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917236134903808
author Shi, Haonan
Shuai, Dehua
Wang, Liming
Liu, Xiyang
Tian, Long
author_facet Shi, Haonan
Shuai, Dehua
Wang, Liming
Liu, Xiyang
Tian, Long
contents Time series forecasting in specialized domains is often constrained by limited data availability, where conventional models typically require large-scale datasets to effectively capture underlying temporal dynamics. To tackle this few-shot challenge, we propose LTSM-DIFF (Large-scale Temporal Sequential Memory with Diffusion), a novel learning framework that integrates the expressive power of large language models with the generative capability of diffusion models. Specifically, the LTSM module is fine-tuned and employed as a temporal memory mechanism, extracting rich sequential representations even under data-scarce conditions. These representations are then utilized as conditional guidance for a joint probability diffusion process, enabling refined modeling of complex temporal patterns. This design allows knowledge transfer from the language domain to time series tasks, substantially enhancing both generalization and robustness. Extensive experiments across diverse benchmarks demonstrate that LTSM-DIFF consistently achieves state-of-the-art performance in data-rich scenarios, while also delivering significant improvements in few-shot forecasting. Our work establishes a new paradigm for time series analysis under data scarcity.
format Preprint
id arxiv_https___arxiv_org_abs_2602_00040
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Enhancing few-shot time series forecasting with LLM-guided diffusion
Shi, Haonan
Shuai, Dehua
Wang, Liming
Liu, Xiyang
Tian, Long
Machine Learning
Artificial Intelligence
Time series forecasting in specialized domains is often constrained by limited data availability, where conventional models typically require large-scale datasets to effectively capture underlying temporal dynamics. To tackle this few-shot challenge, we propose LTSM-DIFF (Large-scale Temporal Sequential Memory with Diffusion), a novel learning framework that integrates the expressive power of large language models with the generative capability of diffusion models. Specifically, the LTSM module is fine-tuned and employed as a temporal memory mechanism, extracting rich sequential representations even under data-scarce conditions. These representations are then utilized as conditional guidance for a joint probability diffusion process, enabling refined modeling of complex temporal patterns. This design allows knowledge transfer from the language domain to time series tasks, substantially enhancing both generalization and robustness. Extensive experiments across diverse benchmarks demonstrate that LTSM-DIFF consistently achieves state-of-the-art performance in data-rich scenarios, while also delivering significant improvements in few-shot forecasting. Our work establishes a new paradigm for time series analysis under data scarcity.
title Enhancing few-shot time series forecasting with LLM-guided diffusion
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2602.00040