UniTS: A unified multi-task time series model

S Gao, T Koker, O Queen… - Advances in …, 2025‏ - proceedings.neurips.cc
Although pre-trained transformers and reprogrammed text-based LLMs have shown strong
performance on time series tasks, the best-performing architectures vary widely across …

Cyclenet: enhancing time series forecasting through modeling periodic patterns

S Lin, W Lin, X Hu, W Wu, R Mo… - Advances in Neural …, 2025‏ - proceedings.neurips.cc
The stable periodic patterns present in time series data serve as the foundation for
conducting long-horizon forecasts. In this paper, we pioneer the exploration of explicitly …

MGSFformer: A multi-granularity spatiotemporal fusion transformer for air quality prediction

C Yu, F Wang, Y Wang, Z Shao, T Sun, D Yao, Y Xu - Information Fusion, 2025‏ - Elsevier
Air quality spatiotemporal prediction can provide technical support for environmental
governance and sustainable city development. As a classic multi-source spatiotemporal …

Time-moe: Billion-scale time series foundation models with mixture of experts

X Shi, S Wang, Y Nie, D Li, Z Ye, Q Wen… - arxiv preprint arxiv …, 2024‏ - arxiv.org
Deep learning for time series forecasting has seen significant advancements over the past
decades. However, despite the success of large-scale pre-training in language and vision …

PGN: The RNN's New Successor is Effective for Long-Range Time Series Forecasting

Y Jia, Y Lin, J Yu, S Wang, T Liu… - Advances in Neural …, 2025‏ - proceedings.neurips.cc
Due to the recurrent structure of RNN, the long information propagation path poses
limitations in capturing long-term dependencies, gradient explosion/vanishing issues, and …

Channel-aware low-rank adaptation in time series forecasting

T Nie, Y Mei, G Qin, J Sun, W Ma - Proceedings of the 33rd ACM …, 2024‏ - dl.acm.org
The balance between model capacity and generalization has been a key focus of recent
discussions in long-term time series forecasting. Two representative channel strategies are …

Not all frequencies are created equal: towards a dynamic fusion of frequencies in time-series forecasting

X Zhang, S Zhao, Z Song, H Guo, J Zhang… - Proceedings of the …, 2024‏ - dl.acm.org
Long-term time series forecasting is a long-standing challenge in various applications. A
central issue in time series forecasting is that methods should expressively capture long …

A Comprehensive Survey of Time Series Forecasting: Architectural Diversity and Open Challenges

J Kim, H Kim, HG Kim, D Lee, S Yoon - arxiv preprint arxiv:2411.05793, 2024‏ - arxiv.org
Time series forecasting is a critical task that provides key information for decision-making
across various fields. Recently, various fundamental deep learning architectures such as …

How Sophisticated Are Neural Networks Needed to Predict Long-Term Nonadiabatic Dynamics?

H Zeng, Y Kou, X Sun - Journal of Chemical Theory and …, 2024‏ - ACS Publications
Nonadiabatic dynamics is key for understanding solar energy conversion and
photochemical processes in condensed phases. This often involves the non-Markovian …

[HTML][HTML] Deep Learning Models for PV Power Forecasting

J Yu, X Li, L Yang, L Li, Z Huang, K Shen, X Yang… - Energies, 2024‏ - mdpi.com
Accurate forecasting of photovoltaic (PV) power is essential for grid scheduling and energy
management. In recent years, deep learning technology has made significant progress in …