将扩散模型融入LLM，提升长时间序列预测的鲁棒性。
Distribution-Aware Diffusion-LLM for Robust Ultra-Long-Term Time Series Forecasting
- 用条件扩散模型增强LLM的分布建模能力
- 在6个数据集上实现超长周期与少样本预测领先效果
- 适合需要高可靠性的长期时间序列场景
时间序列预测是机器学习的基础任务。近期研究探索使用大语言模型（LLMs）因其强大的泛化、模式识别及零样本或少样本能力。尽管适合长上下文学习，但LLMs在多模态设置中面临挑战：缺乏对非文本数据的校准概率建模，且难以对齐异构表示。为此，我们提出新框架Diffusion-LLM，将条件扩散模型嵌入基于LLM的预测流程。该联合设计使模型能学习未来数据的条件分布，同时在共享潜在空间中改善语义对齐。我们在六个长期预测基准（包括ETT、Weather和ECL）上评估了Diffusion-LLM，结果表明其持续优于现有基于LLM的基线，在超长周期和少样本预测中取得显著提升，验证了分布感知正则化对增强时间序列LLM鲁棒性和泛化能力的价值。
原文摘要 · Abstract (English)
Time series forecasting is a fundamental machine learning task. Recent work has explored Large Language Models (LLMs) for this purpose due to their strong generalization, pattern recognition, and zero-shot or few-shot capabilities. Despite their suitability for long-context learning, LLMs face challenges in multimodal settings: they lack calibrated probabilistic modeling for non-text data and struggle to align heterogeneous representations. To address these issues, we propose a new framework Diffusion-LLM that integrates a conditional diffusion model into an LLM-based forecasting pipeline. This joint design enables learning the conditional distribution of future data while improving semantic alignment in a shared latent space. We evaluate Diffusion-LLM on six long-term forecasting benchmarks, including ETT, Weather, and ECL. Our method consistently outperforms existing LLM-based baseline, achieving notable gains in ultra-long-term and few-shot forecasting and demonstrating the value of distribution-aware regularization for enhancing robustness and generalization in time series LLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。