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.
From Values to Tokens: An LLM-Driven Framework for Context-aware Time Series Forecasting via Symbolic Discretization
State Key Laboratory of Cognitive Intelligence（认知智能国家重点实验室）；University of Science and Technology of China（中国科学技术大学）；College of Intelligence and Computing（智能科学与计算学院）；iFLYTEK Research（iFLYTEK研究院）
Asymmetric Peak-Aware Loss for Peak-Critical Time Series Forecasting
Department of Transport and Planning, Faculty of Civil Engineering and Geosciences（交通与规划系，土木与地球科学学院）；Mobility Innovation Centre Delft University of Technology（代尔夫特理工大学移动创新中心）；Delft University of Technology（代尔夫特理工大学）
ReLaMix: Residual Latency-Aware Mixing for Delay-Robust Financial Time-Series Forecasting
Lanzhou University（兰州大学）；Brandeis University（品牌斯大学）；Shenzhen University（深圳大学）；Central University of Finance and Economics（中央财经大学）；Central China Normal University（中央师范大学）
ROAR: Retrieval Opportunity-Aware Refinement for Zero-Shot Time Series Forecasting
College of Computer Science and Technology, Jilin University（吉林大学计算机科学与技术学院）；College of Computing and Data Science, Nanyang Technological University（南洋理工大学计算机与数据科学学院）；City St George’s, University of London（伦敦大学城圣乔治学院）；Systems Research Institute, Polish Academy of Sciences（波兰科学院系统研究所）
Learning Time-aware Graph Structures for Spatially Correlated Time Series Forecasting
Southwest Jiaotong University（西南交通大学）；East China Normal University（华东师范大学）；Aalborg University（奥尔堡大学）；University of Electronic Science and Technology of China（电子科技大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）
HRS: Hybrid Representation Framework with Scheduling Awareness for Time Series Forecasting in Crowdsourced Cloud-Edge Platforms
College of Intelligence and Computing, Tianjin University（天津大学智能与计算学院）；Faculty of Digital Economics and Managements, Tianjin University of Finance and Economics（天津财经大学数字经济学与管理学院）；PPIO Cloud (Shanghai) Co., Ltd.（PPIO云（上海）有限公司）
PA-RNet: Perturbation-Aware Residual Network for Robust Multimodal Time Series Forecasting
Institute of Computing Technology, Guangzhou University（广州大学计算机技术学院）；School of Computer Science and Technology, Dalian University of Technology（大连理工大学计算机科学与技术学院）
RAVEN: A Regime-Aware Variable-context Expert Network for Financial Time Series Forecasting
University of Science and Technology of China（科学技术大学）；Shanghai Black Wing Asset Management Co., Ltd.（上海黑翼资产管理有限公司）；College of Sciences, Shanghai University（上海大学理学院）；The Chinese University of Hong Kong（香港中文大学）；Qitan Technology Co., Ltd.（启元科技有限公司）
IndexNet: Timestamp and Variable-Aware Modeling for Time Series Forecasting
Shanghai Academy of AI for Science（上海人工智能科学研究院）；Fudan University（复旦大学）；Tsinghua Shenzhen International Graduate School（清华大学深圳国际研究生院）；Northeastern University（东北大学）
One Global Model, Many Behaviors: Stockout-Aware Feature Engineering and Dynamic Scaling for Multi-Horizon Retail Demand Forecasting with a Cost-Aware Ordering Policy (VN2 Winner Report)
PESD-TSF: A Period-Aware and Explicit Structured Decomposition Framework for Long-Term Time Series Forecasting
School of Computer and Artificial Intelligence（计算机与人工智能学院）；Ludong University（鲁东大学）；School of Computer Science and Technology（计算机科学与技术学院）；Shandong Technology and Business University（山东科技职业大学）
TA-SparseMG: Trend-Aware Sparse Forecasting via Multi-Scale Gating for Long-Term Time Series
School of Mathematical Sciences, Guizhou Normal University（贵州师范大学数学科学学院）；School of Big Data and Computer Science, Guizhou Normal University（贵州师范大学大数据与计算机科学学院）
交通时间序列预测是具有挑战性的，因为来自不同位置的复杂时空动态时间序列通常具有不同的模式;并且对于相同的时间序列，模式可能随时间而变化，例如，在一天中存在显示出更强的时间相关性的某些时段。尽管最近的预测模型，特别是基于深度学习的模型，显示出有希望的结果，但它们受到时空不可知论的影响。这样的时空不可知模型采用共享参数空间，而不管时间序列位置和时间段，并且它们假设时间模式在位置之间是相似的，并且不会在时间上演变，这可能并不总是成立，从而导致次优结果。在这项工作中，我们提出了一个框架，旨在把时空不可知模型时空感知模型。为此，我们将来自不同位置的时间序列编码为随机变量，从中生成特定于位置和随时间变化的模型参数，以更好地捕捉时空动态。我们展示了如何整合的框架与规范的注意，使时空意识的注意。其次，为了补偿时空感知模型参数生成过程中引入的额外开销，提出了一种新的窗口注意机制，将复杂度从二次降低到线性，使得时空感知注意也具有竞争效率.我们在四个交通时间序列数据集上展示了强有力的经验证据，其中所提出的时空感知注意力在准确性和效率方面优于最先进的方法。这是“Towards Spatio-Temporal Aware Traffic Time Series Forecasting”的扩展版本，将出现在ICDE 2022 [1]中，包括其他实验结果。
Frequency-aware forecasting for short-term typhoon gust prediction
School of Artificial Intelligence, Hubei University（湖北大学人工智能学院）；School of Economics and Management, Wuhan University（武汉大学经济与管理学院）；School of Electrical Engineering, Shanghai Jiao Tong University（上海交通大学电气工程学院）
