Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting

By Yunhao Zhang, Junchi Yan. International Conference on Learning Representations, 2023

1,316 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

ModernTCN: A Modern Pure Convolution Structure for General Time Series Analysis (2024, methodology): “Since cross-variable information is also critical in multivariate time series (Zhang & Yan, 2023; Li et al., 2023b), more time series related modiﬁcations are still needed to make the modern 1D convolution more suitable for time series analysis.”
iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (2023, methodology): “Representative (Zhang & Yan, 2023) explicitly captures the cross-time and cross-variate dependency by the renovated attention mechanism and architecture.”
Transformers in Time-Series Analysis: A Tutorial (2022, background): “Transformers have substantially improved long-term and multi-variate time-series forecasting [29], [30].”
TRACE: Grounding Time Series in Context for Multimodal Embedding and Retrieval (2025, unknown): “Transformer-based models leverage self-attention to capture long-range dependencies and flexible temporal dynamics [19, 20, 21, 22, 23, 24, 25, 26, 27, 28].”
Time to Embed: Unlocking Foundation Models for Time Series with Channel Descriptions (2025, unknown): “Inspired by these successes, the time series community has adopted Transformer-based approaches, leading to notable innovations tailored specifically for temporal data (Zhou et al., 2021b, 2022; Ilbert et al., 2024; Wu et al., 2021; Zhang & Yan, 2023).”