Distribution-Aware Diffusion-LLM for Robust Ultra-Long-Term Time Series Forecasting

The authors propose Diffusion-LLM, a framework that integrates a conditional diffusion model into an LLM-based pipeline to address challenges in multimodal time series forecasting. This joint design enables the learning of future data distributions while improving semantic alignment within a shared latent space.

• Integrates a conditional diffusion model with Large Language Models for time series tasks.
• Enhances robustness and generalization through distribution-aware regularization.
• Evaluated on six long-term forecasting benchmarks, including ETT, Weather, and ECL.
• Achieves notable gains in ultra-long-term and few-shot forecasting scenarios.

The method demonstrates the value of distribution-aware regularization for enhancing robustness and generalization in time series LLMs.