{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T18:58:23Z","timestamp":1785437903866,"version":"3.56.0"},"reference-count":61,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2026,4,21]],"date-time":"2026-04-21T00:00:00Z","timestamp":1776729600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2021YFA1000700"],"award-info":[{"award-number":["2021YFA1000700"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["12031008"],"award-info":[{"award-number":["12031008"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"award":["12031008"],"award-info":[{"award-number":["12031008"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Information"],"abstract":"<jats:p>Adaptive graph learning has become a widely adopted paradigm for multivariate time series forecasting when explicit physical topology is unavailable. In these approaches, node embeddings are typically used to construct dense adjacency matrices based on pairwise similarity, implicitly coupling representation learning with relational modeling. However, we observe that under identical training settings but different random initializations, the learned adjacency matrices can vary substantially while predictive performance remains nearly unchanged, indicating that the relational structure is often underdetermined by the forecasting objective. This observation suggests a mismatch between similarity-based structural learning and the forecasting objective. In this work, we revisit node embeddings from a sequence approximation perspective and propose a Diagonal Adaptive Graph (DiAG) module that restricts adaptive learning to diagonal elements. The diagonal coefficients are derived from channel-independent predictions, while off-diagonal interactions are constructed from the similarity of input sequences. This design decouples representation learning from relational modeling, allowing variables to adaptively switch between channel-independent and channel-dependent regimes. Experiments on multiple datasets show that DiAG improves forecasting performance without modifying the channel-independent backbones. These results indicate that channel-dependent forecasting can be achieved as a prediction-driven refinement over channel-independent backbones, without requiring fully learned dense relational structures.<\/jats:p>","DOI":"10.3390\/info17040394","type":"journal-article","created":{"date-parts":[[2026,4,21]],"date-time":"2026-04-21T11:42:45Z","timestamp":1776771765000},"page":"394","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Diagonal Adaptive Graph: Revisiting Channel Dependency in Multivariate Time Series Forecasting"],"prefix":"10.3390","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-6716-3480","authenticated-orcid":false,"given":"Xiang","family":"Li","sequence":"first","affiliation":[{"name":"School of Mathematics, Shandong University, Jinan 250100, China"},{"name":"Data Science Institute, Shandong University, Jinan 250100, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanping","family":"Zheng","sequence":"additional","affiliation":[{"name":"Gaoling School of Artificial Intelligence, Renmin University of China, Beijing 100872, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhewei","family":"Wei","sequence":"additional","affiliation":[{"name":"Gaoling School of Artificial Intelligence, Renmin University of China, Beijing 100872, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,21]]},"reference":[{"key":"ref_1","unstructured":"Luo, X., Zhu, C., Zhang, D., and Li, Q. (2023). Stg4traffic: A survey and benchmark of spatial-temporal graph neural networks for traffic prediction. arXiv."},{"key":"ref_2","unstructured":"Kipf, T.N., and Welling, M. (2016). Semi-supervised classification with graph convolutional networks. arXiv."},{"key":"ref_3","unstructured":"Veli\u010dkovi\u0107, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., and Bengio, Y. (2017). Graph attention networks. arXiv."},{"key":"ref_4","unstructured":"Xu, Z., Zeng, A., and Xu, Q. (2023). FITS: Modeling Time Series with 10k Parameters. arXiv."},{"key":"ref_5","unstructured":"Wang, H., Pan, L., Shen, Y., Chen, Z., Yang, D., Yang, Y., Zhang, S., Liu, X., Li, H., and Tao, D. (2024). FreDF: Learning to Forecast in the Frequency Domain. arXiv."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1007\/s11280-025-01328-0","article-title":"DSTAN: Attention-enhanced dynamic spatial-temporal network for traffic forecasting","volume":"28","author":"Luo","year":"2025","journal-title":"World Wide Web"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"102293","DOI":"10.1016\/j.inffus.2024.102293","article-title":"Spatiotemporal Fusion Transformer for large-scale traffic forecasting","volume":"107","author":"Wang","year":"2024","journal-title":"Inf. Fusion"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"24326","DOI":"10.52202\/079017-0766","article-title":"Structured Matrix Basis for Multivariate Time Series Forecasting with Interpretable Dynamics","volume":"37","author":"Chen","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"11121","DOI":"10.1609\/aaai.v37i9.26317","article-title":"Are transformers effective for time series forecasting?","volume":"Volume 37","author":"Zeng","year":"2023","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"ref_10","unstructured":"Nie, Y., Nguyen, N.H., Sinthong, P., and Kalagnanam, J. (2022). A Time Series is Worth 64 Words: Long-term Forecasting with Transformers. arXiv."},{"key":"ref_11","unstructured":"Dai, T., Wu, B., Liu, P., Li, N., Bao, J., Jiang, Y., and Xia, S.T. (2024, January 7\u201311). Periodicity decoupling framework for long-term series forecasting. Proceedings of the Twelfth International Conference on Learning Representations, Vienna, Austria."},{"key":"ref_12","unstructured":"Zhang, Y., and Yan, J. (2023, January 1\u20135). Crossformer: Transformer utilizing cross-dimension dependency for multivariate time series forecasting. Proceedings of the Eleventh International Conference on Learning Representations, Kigali, Rwanda."},{"key":"ref_13","unstructured":"Liu, Y., Hu, T., Zhang, H., Wu, H., Wang, S., Ma, L., and Long, M. (2024, January 7\u201311). iTransformer: Inverted Transformers Are Effective for Time Series Forecasting. Proceedings of the Twelfth International Conference on Learning Representations, Vienna, Austria."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"114180","DOI":"10.52202\/079017-3627","article-title":"Are self-attentions effective for time series forecasting?","volume":"37","author":"Kim","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"130635","DOI":"10.52202\/079017-4152","article-title":"From similarity to superiority: Channel clustering for time series forecasting","volume":"37","author":"Chen","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Qiu, X., Wu, X., Lin, Y., Guo, C., Hu, J., and Yang, B. (2025, January 3\u20137). Duet: Dual clustering enhanced multivariate time series forecasting. Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1, Toronto, ON, Canada.","DOI":"10.1145\/3690624.3709325"},{"key":"ref_17","unstructured":"Hu, Y., Zhang, G., Liu, P., Lan, D., Li, N., Cheng, D., Dai, T., Xia, S.T., and Pan, S. (2025, January 13\u201319). TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecasting. Proceedings of the International Conference on Machine Learning, PMLR, Vancouver, BC, Canada."},{"key":"ref_18","unstructured":"Lin, S., Lin, W., Wu, W., Chen, H., and Yang, J. (2024, January 21\u201327). SparseTSF: Modeling Long-term Time Series Forecasting with *1k* Parameters. Proceedings of the International Conference on Machine Learning, PMLR, Vienna, Austria."},{"key":"ref_19","first-page":"57818","article-title":"Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling","volume":"235","author":"Yu","year":"2024","journal-title":"Proc. Mach. Learn. Res."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Xue, J., Tan, R., Ma, J., and Ukkusuri, S.V. (2025). Data Mining in Transportation Networks with Graph Neural Networks: A Review and Outlook. arXiv.","DOI":"10.1007\/s42421-025-00124-6"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1007\/s42421-025-00124-6","article-title":"Data Science in Transportation Networks with Graph Neural Networks: A Review and Outlook","volume":"7","author":"Xue","year":"2025","journal-title":"Data Sci. Transp."},{"key":"ref_22","unstructured":"Li, Y., Yu, R., Shahabi, C., and Liu, Y. (2017). Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. arXiv."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Wu, Z., Pan, S., Long, G., Jiang, J., and Zhang, C. (2019). Graph wavenet for deep spatial-temporal graph modeling. arXiv.","DOI":"10.24963\/ijcai.2019\/264"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"107382","DOI":"10.1016\/j.neunet.2025.107382","article-title":"Bilinear Spatiotemporal Fusion Network: An efficient approach for traffic flow prediction","volume":"187","author":"Chen","year":"2025","journal-title":"Neural Netw."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"106805","DOI":"10.1016\/j.neunet.2024.106805","article-title":"Dynamic meta-graph convolutional recurrent network for heterogeneous spatiotemporal graph forecasting","volume":"181","author":"Guo","year":"2025","journal-title":"Neural Netw."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Tang, J., Wei, W., Xia, L., and Huang, C. (2024, January 21\u201325). EasyST: A Simple Framework for Spatio-Temporal Prediction. Proceedings of the 33rd ACM International Conference on Information and Knowledge Management, Boise, ID, USA.","DOI":"10.1145\/3627673.3679749"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"2517","DOI":"10.1109\/TBDATA.2025.3544131","article-title":"LightST: A simplifying spatio-temporal graph neural network for traffic flow forecasting","volume":"11","author":"Hu","year":"2025","journal-title":"IEEE Trans. Big Data"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Fang, Y., Liang, Y., Hui, B., Shao, Z., Deng, L., Liu, X., Jiang, X., and Zheng, K. (2025, January 3\u20137). Efficient large-scale traffic forecasting with transformers: A spatial data management perspective. Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1, Toronto, ON, Canada.","DOI":"10.1145\/3690624.3709177"},{"key":"ref_29","unstructured":"Yu, C., Xie, X., Huang, Y., and Qiu, C. (2025). FlowDistill: Scalable Traffic Flow Prediction via Distillation from LLMs. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"4846","DOI":"10.1109\/TKDE.2025.3570705","article-title":"ST-LLM+: Graph Enhanced Spatio-Temporal Large Language Models for Traffic Prediction","volume":"37","author":"Liu","year":"2025","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"225","DOI":"10.1109\/TAI.2025.3571378","article-title":"Spatio-Temporal Mamba Dynamic Graph Convolutional Recurrent Network for Traffic Prediction","volume":"7","author":"Zhang","year":"2025","journal-title":"IEEE Trans. Artif. Intell."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"102872","DOI":"10.1016\/j.inffus.2024.102872","article-title":"ST-MambaSync: Complement the power of Mamba and Transformer fusion for less computational cost in spatial\u2013temporal traffic forecasting","volume":"117","author":"Shao","year":"2025","journal-title":"Inf. Fusion"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Yuan, Y., Ding, J., Feng, J., Jin, D., and Li, Y. (2024, January 25\u201329). Unist: A prompt-empowered universal model for urban spatio-temporal prediction. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Barcelona, Spain.","DOI":"10.1145\/3637528.3671662"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"102542","DOI":"10.1016\/j.inffus.2024.102542","article-title":"KaTaGCN: Knowledge-augmented and time-aware graph convolutional network for efficient traffic forecasting","volume":"111","author":"Wang","year":"2024","journal-title":"Inf. Fusion"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Dong, Z., Jiang, R., Gao, H., Liu, H., Deng, J., Wen, Q., and Song, X. (2024, January 25\u201329). Heterogeneity-informed meta-parameter learning for spatiotemporal time series forecasting. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Barcelona, Spain.","DOI":"10.1145\/3637528.3671961"},{"key":"ref_36","first-page":"1","article-title":"Traffic state prediction based on adaptive spatial-temporal information","volume":"14","author":"An","year":"2024","journal-title":"Authorea Prepr."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"4776","DOI":"10.1109\/TKDE.2025.3571983","article-title":"QSTGNN: Quaternion Spatio-Temporal Graph Neural Networks","volume":"37","author":"Liu","year":"2025","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Zhu, G., Hou, H., Wang, P., Yuan, C., and Huang, Y. (2023). STSD: Modeling Spatial Temporal Staticity and Dynamicity in Traffic Forecasting. Proceedings of the 2023 IEEE International Conference on Data Mining (ICDM), IEEE.","DOI":"10.1109\/ICDM58522.2023.00209"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Wu, M., Lin, Y., Jiang, T., and Weng, W. (2025). MHGNet: Multi-Heterogeneous Graph Neural Network for Traffic Prediction. Proceedings of the ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE.","DOI":"10.1109\/ICASSP49660.2025.10890207"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Chen, J., Shao, Q., Chen, D., and Yu, W. (2025, January 3\u20137). Decoupling Spatio-Temporal Prediction: When Lightweight Large Models Meet Adaptive Hypergraphs. Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2, Toronto, ON, Canada.","DOI":"10.1145\/3711896.3736904"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"122363","DOI":"10.1016\/j.ins.2025.122363","article-title":"STIFF: Spatio-Temporal Feature Fusion Model with Interactive Learning for Traffic Flow Prediction","volume":"718","author":"Qiao","year":"2025","journal-title":"Inf. Sci."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"27379","DOI":"10.1109\/JIOT.2025.3562861","article-title":"Dynamic Graph Convolution and Spatio-Temporal Self-Attention Network for Traffic Flow Prediction","volume":"12","author":"Liu","year":"2025","journal-title":"IEEE Internet Things J."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"112586","DOI":"10.1016\/j.knosys.2024.112586","article-title":"Dynamic multi-scale spatial-temporal graph convolutional network for traffic flow prediction","volume":"305","author":"Gao","year":"2024","journal-title":"Knowl.-Based Syst."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"113178","DOI":"10.1016\/j.asoc.2025.113178","article-title":"MGC-DMF: A traffic flow forecasting method based on multi-graph spatio-temporal convolution and dynamic metric fusion with multi-source basic information","volume":"177","author":"Feng","year":"2025","journal-title":"Appl. Soft Comput."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Zou, D., Wang, S., Li, X., Peng, H., Wang, Y., Liu, C., Sheng, K., and Zhang, B. (2024, January 4\u20138). Multispans: A multi-range spatial-temporal transformer network for traffic forecast via structural entropy optimization. Proceedings of the 17th ACM International Conference on Web Search and Data Mining, Merida, Mexico.","DOI":"10.1145\/3616855.3635820"},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"112892","DOI":"10.1016\/j.asoc.2025.112892","article-title":"SDSINet: A spatiotemporal dual-scale interaction network for traffic prediction","volume":"173","author":"Yang","year":"2025","journal-title":"Appl. Soft Comput."},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"113147","DOI":"10.1016\/j.knosys.2025.113147","article-title":"TVC Former: A transformer-based long-term multivariate time series forecasting method using time-variable coupling correlation graph","volume":"314","author":"Liu","year":"2025","journal-title":"Knowl.-Based Syst."},{"key":"ref_48","first-page":"64145","article-title":"Softs: Efficient multivariate time series forecasting with series-core fusion","volume":"37","author":"Han","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_49","unstructured":"Wu, H., Hu, T., Liu, Y., Zhou, H., Wang, J., and Long, M. (2022). TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis. arXiv."},{"key":"ref_50","unstructured":"Wang, Y., Wu, H., Ma, Y., Fang, Y., Zhang, Z., Liu, Y., Wang, S., Ye, Z., Xiang, Y., and Wang, J. (2025). Accuracy law for the future of deep time series forecasting. arXiv."},{"key":"ref_51","first-page":"83314","article-title":"Scaling law for time series forecasting","volume":"37","author":"Shi","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_52","unstructured":"Wang, F., Li, Y., Shao, Z., Yu, C., Fu, Y., An, Z., Xu, Y., and Cheng, X. (2025). Aries: Relation assessment and model recommendation for deep time series forecasting. arXiv."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"2059","DOI":"10.1109\/TAI.2025.3610393","article-title":"Air quality prediction with a meteorology-guided modality-decoupled spatio-temporal network","volume":"7","author":"Yin","year":"2025","journal-title":"IEEE Trans. Artif. Intell."},{"key":"ref_54","unstructured":"Shi, J., Shirali, A., Jin, B., Zhou, S., Hu, W., Rangaraj, R., Wang, S., Han, J., Wang, Z., and Lall, U. (2025). Deep learning and foundation models for weather prediction: A survey. arXiv."},{"key":"ref_55","first-page":"17804","article-title":"Adaptive graph convolutional recurrent network for traffic forecasting","volume":"33","author":"Bai","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_56","first-page":"1","article-title":"Dynamic graph convolutional recurrent network for traffic prediction: Benchmark and solution","volume":"17","author":"Li","year":"2023","journal-title":"ACM Trans. Knowl. Discov. Data"},{"key":"ref_57","doi-asserted-by":"crossref","unstructured":"Wu, Z., Pan, S., Long, G., Jiang, J., Chang, X., and Zhang, C. (2020, January 23\u201327). Connecting the dots: Multivariate time series forecasting with graph neural networks. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Virtual Event.","DOI":"10.1145\/3394486.3403118"},{"key":"ref_58","doi-asserted-by":"crossref","unstructured":"Shao, Z., Zhang, Z., Wei, W., Wang, F., Xu, Y., Cao, X., and Jensen, C.S. (2022). Decoupled dynamic spatial-temporal graph neural network for traffic forecasting. arXiv.","DOI":"10.14778\/3551793.3551827"},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"1081","DOI":"10.14778\/3641204.3641217","article-title":"Bigst: Linear complexity spatio-temporal graph neural network for traffic forecasting on large-scale road networks","volume":"17","author":"Han","year":"2024","journal-title":"Proc. VLDB Endow."},{"key":"ref_60","doi-asserted-by":"crossref","unstructured":"Wills, P., and Meyer, F.G. (2020). Metrics for graph comparison: A practitioner\u2019s guide. PLoS ONE, 15.","DOI":"10.1371\/journal.pone.0228728"},{"key":"ref_61","doi-asserted-by":"crossref","unstructured":"Koutra, D., Vogelstein, J.T., and Faloutsos, C. (2013). Deltacon: A principled massive-graph similarity function. Proceedings of the 2013 SIAM International Conference on Data Mining, SIAM.","DOI":"10.1137\/1.9781611972832.18"}],"container-title":["Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2078-2489\/17\/4\/394\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,21]],"date-time":"2026-04-21T12:10:53Z","timestamp":1776773453000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2078-2489\/17\/4\/394"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,21]]},"references-count":61,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["info17040394"],"URL":"https:\/\/doi.org\/10.3390\/info17040394","relation":{},"ISSN":["2078-2489"],"issn-type":[{"value":"2078-2489","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,21]]}}}