{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T14:09:18Z","timestamp":1779286158965,"version":"3.51.4"},"reference-count":45,"publisher":"Association for Computing Machinery (ACM)","issue":"5","license":[{"start":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T00:00:00Z","timestamp":1779235200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"Natural Science Foundation of Guangdong Province, China","award":["2026A1515010176"],"award-info":[{"award-number":["2026A1515010176"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Knowl. Discov. Data"],"published-print":{"date-parts":[[2026,6,30]]},"abstract":"<jats:p>The task of dynamic graph link prediction is to forecast the evolution of complex systems. Empirical observations reveal that interactions within these systems exhibit an Entangled Spatio-Temporal Pattern, which manifests through three interrelated phenomena, namely Latent High-Order Bridges, Multi-Frequency Temporal Dynamics, and Spatio-Temporal Entanglement, with stronger structural ties facilitating tolerance for longer temporal gaps. However, limited by computationally prohibitive multi-hop sampling or inefficient long-sequence modeling, existing methods struggle to capture this complex pattern. Inspired by State-Space Models (SSMs) like Mamba for efficient long-range modeling yet aiming to address their native agnosticism to structural and multi-frequency dynamics, we propose a framework named DyGHydra, which couples a tailored Continuous-Time Hierarchical Mamba (CT-HMamba) backbone with a multi-hop structural encoder. The framework first employs the multi-hop structural encoder to reveal latent high-order interactions, extracting interaction-level cross-hop features. Subsequently, the CT-HMamba backbone utilizes these features to address multi-frequency dynamics through a hierarchical architecture, decomposing interaction history to simultaneously model high-frequency bursts and long-term trends. To capture the spatio-temporal entanglement, CT-HMamba further tailors its core state-space mechanism to be co-driven by physical time and structural context. Specifically, physical time governs the state transition decay to reflect temporal forgetting, while structural context modulates the input-output projections to prioritize topologically significant events. Extensive experiments on eleven real-world datasets show that DyGHydra achieves state-of-the-art performance across most settings for both transductive and inductive link prediction, validating its effectiveness in modeling complex temporal dynamics with superior efficiency.<\/jats:p>","DOI":"10.1145\/3807958","type":"journal-article","created":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T15:39:57Z","timestamp":1777390797000},"page":"1-28","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["DyGHydra: A Hierarchical State-Space Model with Time Dynamics and Interactive-Relational Selectivity for Link Prediction"],"prefix":"10.1145","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-1569-1441","authenticated-orcid":false,"given":"Yueqi","family":"Guo","sequence":"first","affiliation":[{"name":"School of Computer Science, South China Normal University, Guangzhou, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-6210-8038","authenticated-orcid":false,"given":"Haojie","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Computer Science, South China Normal University, Guangzhou, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0727-4744","authenticated-orcid":false,"given":"Weihao","family":"Yu","sequence":"additional","affiliation":[{"name":"China Telecom Research Institute Guangzhou, Guangzhou, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2285-5248","authenticated-orcid":false,"given":"Jin","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Computer Science, South China Normal University, Guangzhou, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2026,5,20]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41562-020-01024-1"},{"key":"e_1_3_1_3_2","doi-asserted-by":"publisher","DOI":"10.1038\/nature03459"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.5555\/3327757.3327764"},{"key":"e_1_3_1_5_2","first-page":"301","volume-title":"Proceedings of the ACM International Conference on Information and Knowledge Management","author":"Chen Xi","year":"2024","unstructured":"Xi Chen, Yun Xiong, Siwei Zhang, Jiawei Zhang, Yao Zhang, Shiyang Zhou, Xixi Wu, Mingyang Zhang, Tengfei Liu, and Weiqiang Wang. 2024. DTFormer: A transformer-based method for discrete-time dynamic graph representation learning. In Proceedings of the ACM International Conference on Information and Knowledge Management, 301\u2013311."},{"key":"e_1_3_1_6_2","doi-asserted-by":"crossref","first-page":"421","DOI":"10.1145\/3637528.3671770","volume-title":"Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Cheng K.","year":"2024","unstructured":"K. Cheng, L. Peng, J. Ye, L. Sun, and B. Du. 2024. Co-neighbor encoding schema: A light-cost structure encoding method for dynamic link prediction. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 421\u2013432."},{"key":"e_1_3_1_7_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Cong W.","year":"2023","unstructured":"W. Cong, S. Zhang, J. Kang, B. Yuan, H. Wu, X. Zhou, H. Tong, and M. Mahdavi. 2023. Do we really need complicated model architectures for temporal networks? In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_8_2","first-page":"1","article-title":"DyGMamba: Efficiently modeling long-term temporal dependency on continuous-time dynamic graphs with state space models","author":"Ding Zifeng","year":"2025","unstructured":"Zifeng Ding, Yifeng Li, Yuan He, Antonio Norelli, Jingcheng Wu, Volker Tresp, Michael M. Bronstein, and Yunpu Ma. 2025. DyGMamba: Efficiently modeling long-term temporal dependency on continuous-time dynamic graphs with state space models. Transactions on Machine Learning Research 2025 (May 2025), 1\u201331.","journal-title":"Transactions on Machine Learning Research"},{"key":"e_1_3_1_9_2","volume-title":"\u00dcber Das Ged\u00e4chtnis: Untersuchungen zur Experimentellen Psychologie","author":"Ebbinghaus Hermann","year":"1885","unstructured":"Hermann Ebbinghaus. 1885. \u00dcber Das Ged\u00e4chtnis: Untersuchungen zur Experimentellen Psychologie. Duncker & Humblot."},{"key":"e_1_3_1_10_2","first-page":"433","volume-title":"Proceedings of the ACM Conference on Information and Knowledge Management","author":"Fan Ziwei","year":"2021","unstructured":"Ziwei Fan, Zhiwei Liu, Jiawei Zhang, Yun Xiong, Lei Zheng, and Philip S. Yu. 2021. Continuous-time sequential recommendation with temporal graph collaborative transformer. In Proceedings of the ACM Conference on Information and Knowledge Management, 433\u2013442."},{"key":"e_1_3_1_11_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2019.06.024"},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.2307\/202051"},{"key":"e_1_3_1_13_2","volume-title":"Proceedings of the 1st Conference on Language Modeling","author":"Gu Albert","year":"2024","unstructured":"Albert Gu and Tri Dao. 2024. Mamba: Linear-time sequence modeling with selective state spaces. In Proceedings of the 1st Conference on Language Modeling."},{"key":"e_1_3_1_14_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Gu Albert","year":"2022","unstructured":"Albert Gu, Karan Goel, and Christopher Re. 2022. Efficiently modeling long sequences with structured state spaces. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_15_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2024.111590"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.01.039"},{"key":"e_1_3_1_17_2","first-page":"22765","volume-title":"Proceedings of the Conference on Neural Information Processing Systems","author":"Huang Xuanwen","year":"2022","unstructured":"Xuanwen Huang, Yang Yang, Yang Wang, Chunping Wang, Zhisheng Zhang, Jiarong Xu, Lei Chen, and Michalis Vazirgiannis. 2022. DGraph: A large-scale financial dataset for graph anomaly detection. In Proceedings of the Conference on Neural Information Processing Systems, 22765\u201322777."},{"key":"e_1_3_1_18_2","first-page":"16177","volume-title":"Advances in Neural Information Processing Systems","volume":"33","author":"Huang Zijie","year":"2020","unstructured":"Zijie Huang, Yizhou Sun, and Wei Wang. 2020. Learning continuous system dynamics from irregularly-sampled partial observations. In Advances in Neural Information Processing Systems, Vol. 33, 16177\u201316187."},{"key":"e_1_3_1_19_2","doi-asserted-by":"crossref","unstructured":"Ke Jin Weihao Yu Yingchao Long Nanhui Lai and Jin Huang. 2025. A diffusion multi-interest framework for cross-domain recommendation. Expert Systems with Applications 294 (2025) 128738.","DOI":"10.1016\/j.eswa.2025.128738"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330895"},{"key":"e_1_3_1_21_2","doi-asserted-by":"crossref","unstructured":"Jure Leskovec and Rok Sosi\u010d. 2016. SNAP: A general-purpose network analysis and graph-mining library. ACM Transactions on Intelligent Systems and Technology 8 1 (July 2016) 1\u201320.","DOI":"10.1145\/2898361"},{"key":"e_1_3_1_22_2","first-page":"5363","volume-title":"Proceedings of the Conference on Neural Information Processing Systems","author":"Li Pan","year":"2020","unstructured":"Pan Li, Yanbang Wang, Hongwei Wang, and Jure Leskovec. 2020. Distance encoding: design provably more powerful neural networks for graph representation learning. In Proceedings of the Conference on Neural Information Processing Systems, 5363\u20135376."},{"key":"e_1_3_1_23_2","first-page":"322","volume-title":"Proceedings of the IEEE International Conference on Data Mining (ICDM)","author":"Li Xiaohan","year":"2020","unstructured":"Xiaohan Li, Mengqi Zhang, Shu Wu, Zheng Liu, Liang Wang, and Philip S. Yu. 2020. Dynamic graph collaborative filtering. In Proceedings of the IEEE International Conference on Data Mining (ICDM), 322\u2013331."},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5984"},{"key":"e_1_3_1_25_2","volume-title":"Proceedings of the 1st International Workshop on Deep Learning on Graphs: Methodologies and Applications","author":"Poli Michael","year":"2020","unstructured":"Michael Poli, Stefano Massaroli, Junyoung Park, Atsushi Yamashita, Hajime Asama, and Jinkyoo Park. 2020. Graph neural ordinary differential equations. In Proceedings of the 1st International Workshop on Deep Learning on Graphs: Methodologies and Applications. Association for the Advancement of Artificial Intelligence."},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1145\/3625820"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1111\/j.1475-6811.2010.01310.x"},{"key":"e_1_3_1_28_2","volume-title":"Proceedings of the International Conference on Machine Learning on Workshop Graph Representation Learning","author":"Rossi Emanuele","year":"2020","unstructured":"Emanuele Rossi, Ben Chamberlain, Fabrizio Frasca, Davide Eynard, Federico Monti, and Michael Bronstein. 2020. Temporal graph networks for deep learning on dynamic graphs. In Proceedings of the International Conference on Machine Learning on Workshop Graph Representation Learning."},{"key":"e_1_3_1_29_2","doi-asserted-by":"crossref","first-page":"519","DOI":"10.1145\/3336191.3371845","volume-title":"Proceedings of the International Conference on Web Search & Data Mining","author":"Sankar Aravind","year":"2020","unstructured":"Aravind Sankar, Yanhong Wu, Liang Gou, Wei Zhang, and Hao Yang. 2020. DySAT: Deep neural representation learning on dynamic graphs via self-attention networks. In Proceedings of the International Conference on Web Search & Data Mining, 519\u2013527."},{"key":"e_1_3_1_30_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Smith Jimmy T. H.","year":"2023","unstructured":"Jimmy T. H. Smith, Andrew Warrington, and Scott Linderman. 2023. Simplified state space layers for sequence modeling. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1145\/3289600.3290989"},{"key":"e_1_3_1_32_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Tian Yuxing","year":"2024","unstructured":"Yuxing Tian, Yiyan Qi, and Fan Guo. 2024. FreeDyG: Frequency enhanced continuous-time dynamic graph model for link prediction. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_33_2","unstructured":"Ilya O. Tolstikhin Neil Houlsby Alexander Kolesnikov Lucas Beyer Xiaohua Zhai Thomas Unterthiner Jessica Yung Andreas Steiner Daniel Keysers Jakob Uszkoreit et al. 2021. MLP-mixer: An all-MLP architecture for vision. In Advances in Neural Information Processing Systems Vol. 34 24261\u201324272."},{"key":"e_1_3_1_34_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Trivedi Rakshit","year":"2019","unstructured":"Rakshit Trivedi, Mehrdad Farajtabar, Prasenjeet Biswal, and Hongyuan Zha. 2019. DyRep: learning representations over dynamic graphs. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_35_2","unstructured":"Lu Wang Xiaofu Chang Shuang Li Yunfei Chu Hui Li Wei Zhang Xiaofeng He Le Song Jingren Zhou and Hongxia Yang. 2021. TCL: Transformer-based dynamic graph modelling via contrastive learning. arXiv:2105.07944. Retrieved from https:\/\/arxiv.org\/abs\/2105.07944"},{"key":"e_1_3_1_36_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Wang Y.","year":"2021","unstructured":"Y. Wang, Y.-Y. Chang, Y. Liu, J. Leskovec, and P. Li. 2021. Inductive representation learning in temporal networks via causal anonymous walks. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_37_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Wu Haixu","year":"2023","unstructured":"Haixu Wu, Tengge Hu, Yong Liu, Hang Zhou, Jianmin Wang, and Mingsheng Long. 2023. TimesNet: Temporal 2D-variation modeling for general time series analysis. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_38_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Xu Da","year":"2020","unstructured":"Da Xu, Chuanwei Ruan, Evren Korpeoglu, Sushant Kumar, and Kannan Achan. 2020. Inductive representation learning on temporal graphs. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_1_39_2","first-page":"2358","volume-title":"Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"You Jiaxuan","year":"2022","unstructured":"Jiaxuan You, Tianyu Du, and Jure Leskovec. 2022. ROLAND: Graph learning framework for dynamic graphs. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2358\u20132366."},{"issue":"4","key":"e_1_3_1_40_2","first-page":"1475","article-title":"Continuous-time user preference modelling for temporal sets prediction","volume":"36","author":"Yu Le","year":"2023","unstructured":"Le Yu, Zihang Liu, Leilei Sun, Bowen Du, Chuanren Liu, and Weifeng Lv. 2023. Continuous-time user preference modelling for temporal sets prediction. IEEE Transactions on Knowledge and Data Engineering 36, 4 (2023), 1475\u20131488.","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"e_1_3_1_41_2","doi-asserted-by":"crossref","unstructured":"Le Yu Leilei Sun Bowen Du and Weifeng Lv. 2023. Towards better dynamic graph learning: New architecture and unified library. In Advances in Neural Information Processing Systems Vol. 36 67686\u201367700.","DOI":"10.52202\/075280-2960"},{"key":"e_1_3_1_42_2","first-page":"1902","volume-title":"Proceedings of the ACM Web Conference","author":"Yu Le","year":"2022","unstructured":"Le Yu, Guanghui Wu, Leilei Sun, Bowen Du, and Weifeng Lv. 2022. Element-guided temporal graph representation learning for temporal sets prediction. In Proceedings of the ACM Web Conference, 1902\u20131913."},{"key":"e_1_3_1_43_2","first-page":"139","volume-title":"Proceedings of the British Machine Vision Conference","author":"Yu Tan","year":"2021","unstructured":"Tan Yu, Xu Li, Yunfeng Cai, Mingming Sun, and Ping Li. 2021. Rethinking token-mixing MLP for MLP-based vision backbone. In Proceedings of the British Machine Vision Conference, 139."},{"key":"e_1_3_1_44_2","unstructured":"Mengqi Zhang Shu Wu Xueli Yu Qiang Liu and Liang Wang. 2022. Dynamic graph neural networks for sequential recommendation. IEEE Transactions on Knowledge and Data Engineering 35 5 (2022) 4741\u20134753."},{"key":"e_1_3_1_45_2","first-page":"27268","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Zhou Tian","year":"2022","unstructured":"Tian Zhou, Ziqing Ma, Qingsong Wen, Xue Wang, Liang Sun, and Rong Jin. 2022. Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting. In Proceedings of the International Conference on Machine Learning, 27268\u201327286."},{"key":"e_1_3_1_46_2","first-page":"1","volume-title":"Proceedings of the International Conference on Algorithms, Computing, and Artificial Intelligence","author":"Zhu Wentai","year":"2022","unstructured":"Wentai Zhu, Ke Ruan, Jin Huang, Jing Xiao, and Weihao Yu. 2022. Dynamic graph representation based on temporal and contextual contrasting. In Proceedings of the International Conference on Algorithms, Computing, and Artificial Intelligence, 1\u20138."}],"container-title":["ACM Transactions on Knowledge Discovery from Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3807958","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T13:31:07Z","timestamp":1779283867000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3807958"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,20]]},"references-count":45,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,6,30]]}},"alternative-id":["10.1145\/3807958"],"URL":"https:\/\/doi.org\/10.1145\/3807958","relation":{},"ISSN":["1556-4681","1556-472X"],"issn-type":[{"value":"1556-4681","type":"print"},{"value":"1556-472X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,20]]},"assertion":[{"value":"2025-12-18","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-27","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-20","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}