---
title: "Speaking Numbers to LLMs: Multi-Wavelet Number Embeddings for Time Series Forecasting"
authors:
  - "Defu Cao"
  - "Zijie Lei"
  - "Muyan Weng"
  - "Jiao Sun"
  - "Yan Liu"
doi: "10.24963/ijcai.2026/464"
url: "https://ijcai.org/proceedings/2026/464"
pdf: "https://ijcai.org/proceedings/2026/0464.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "4170-4178"
track: "Main Track"
keywords:
  - "Natural Language Processing: Language models"
  - "Machine Learning: Time series and data streams"
  - "Natural Language Processing: Embeddings"
---

# Speaking Numbers to LLMs: Multi-Wavelet Number Embeddings for Time Series Forecasting

Defu Cao, Zijie Lei, Muyan Weng, Jiao Sun, Yan Liu

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 4170-4178. DOI: [10.24963/ijcai.2026/464](https://doi.org/10.24963/ijcai.2026/464)

## Abstract

Large language models (LLMs) are attractive for context-aware time series forecasting because they can integrate heterogeneous textual signals, yet their discrete, language-oriented tokenization and embedding interfaces are misaligned with continuous numerical values, often harming numerical ordering and forecasting reliability. We propose TempoWave, a plug-and-play temporal wavelet digit interface that maps each scalar observation into digit-wise embeddings constructed from multi-wavelet, multi-scale coefficients. By directly overriding standard token representations, TempoWave seamlessly exposes both fine-grained local fluctuations and macro global structures in a transformer-compatible form, ensuring that precise numerical formatting, distinct digit identity, and robustness to common normalization operations are maintained throughout the LLM pipeline. Experiments across five context-enriched forecasting benchmarks demonstrate that TempoWave consistently improves LLM-based forecasters over standard numeric tokenization and alternative embedding interfaces, achieving a new state-of-the-art. These results highlight the numeric interface as a key bottleneck and suggest that principled multi-resolution embeddings can better couple LLMs' contextual reasoning with precise forecasting. Our code is available at https://github.com/DC-research/TempoWAVE.