{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T09:33:43Z","timestamp":1776850423366,"version":"3.51.2"},"reference-count":49,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T00:00:00Z","timestamp":1776816000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Key Research and Development Program of the Ministry of Science and Technology of China","award":["2022YFC3800500"],"award-info":[{"award-number":["2022YFC3800500"]}]},{"name":"Science and Technology Research Program of Chongqing Municipal Education Commission","award":["KJQN202301543"],"award-info":[{"award-number":["KJQN202301543"]}]},{"name":"Natural Science Foundation of Chongqing, China","award":["CSTB2024NSCQ-MSX0009"],"award-info":[{"award-number":["CSTB2024NSCQ-MSX0009"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["51878094"],"award-info":[{"award-number":["51878094"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Creative Research Groups in Colleges and Universities of Chongqing, China","award":["CXQT21001"],"award-info":[{"award-number":["CXQT21001"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Accurate traffic flow prediction is central to Intelligent Transportation Systems yet remains difficult due to non-Euclidean spatial structure, long-range propagation, and time-varying delays. However, existing deep learning methods have key limitations, including reliance on fixed temporal partitioning, small-kernel locality that restricts receptive fields, and graph constructions that lack long-range, dynamic correlations. This work presents dual spatio-temporal graph and large-kernel convolutionnetwork (DTG-LKNet), a Transformer-based architecture that packages a large-kernel convolution module with a dual spatio-temporal graph in a unified framework. On the temporal side, Deformable Patch Sampling learns sampling offsets around salient timestamps, and large kernels expand the effective receptive field without deep dilation stacks. On the spatial side, DTG-LKNet fuses a functional-similarity graph with the physical road network topology to represent long-range correlations. Comprehensive experiments on three large-scale benchmarks demonstrate consistent state-of-the-art performance against strong baselines. This article further visualizes the effective receptive field, validating the effectiveness of the large-kernel design.<\/jats:p>","DOI":"10.7717\/peerj-cs.3793","type":"journal-article","created":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T08:33:41Z","timestamp":1776846821000},"page":"e3793","source":"Crossref","is-referenced-by-count":0,"title":["DTG-LKNet: dual spatio-temporal graphs and large-kernel convolutions network for traffic prediction"],"prefix":"10.7717","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-3204-1939","authenticated-orcid":true,"given":"Jiahao","family":"Cao","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1554-4791","authenticated-orcid":true,"given":"Yuan","family":"Tian","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"YangSheng","family":"Long","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Peng","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tong","family":"Xiao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Peng","family":"Ye","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guoqing","family":"Teng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"4443","published-online":{"date-parts":[[2026,4,22]]},"reference":[{"issue":"23","key":"10.7717\/peerj-cs.3793\/ref-1","doi-asserted-by":"publisher","first-page":"11047","DOI":"10.3390\/app142311047","article-title":"Machine learning models for traffic prediction on arterial roads using traffic features and weather information","volume":"14","author":"Abduljabbar","year":"2024","journal-title":"Applied Sciences"},{"issue":"3","key":"10.7717\/peerj-cs.3793\/ref-2","doi-asserted-by":"publisher","first-page":"116898","DOI":"10.1016\/j.chaos.2025.116898","article-title":"Dynamic multi-graph spatio-temporal learning for citywide traffic flow prediction in transportation systems","volume":"199","author":"Ali","year":"2025a","journal-title":"Chaos, Solitons & Fractals"},{"issue":"9","key":"10.7717\/peerj-cs.3793\/ref-3","doi-asserted-by":"publisher","first-page":"14154","DOI":"10.1109\/tits.2025.3540852","article-title":"An attention-driven spatio-temporal deep hybrid neural networks for traffic flow prediction in transportation systems","volume":"26","author":"Ali","year":"2025b","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"issue":"4","key":"10.7717\/peerj-cs.3793\/ref-4","doi-asserted-by":"publisher","first-page":"7252","DOI":"10.1109\/tce.2024.3439719","article-title":"A resource-aware multi-graph neural network for urban traffic flow prediction in multi-access edge computing systems","volume":"70","author":"Ali","year":"2024","journal-title":"IEEE Transactions on Consumer Electronics"},{"issue":"4","key":"10.7717\/peerj-cs.3793\/ref-5","doi-asserted-by":"publisher","first-page":"879","DOI":"10.3390\/electronics12040879","article-title":"Improved graph convolutional network with enriched graph topology representation for skeleton-based action recognition","volume":"12","author":"Alsarhan","year":"2023","journal-title":"Electronics"},{"issue":"3","key":"10.7717\/peerj-cs.3793\/ref-6","first-page":"17804","article-title":"Adaptive graph convolutional recurrent network for traffic forecasting","volume":"33","author":"Bai","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"5\u20136","key":"10.7717\/peerj-cs.3793\/ref-7","doi-asserted-by":"publisher","first-page":"909","DOI":"10.1007\/s00779-019-01212-5","article-title":"A two-stage road traffic congestion prediction and resource dispatching toward a self-organizing traffic control system","volume":"23","author":"Bouyahia","year":"2019","journal-title":"Personal and Ubiquitous Computing"},{"issue":"7","key":"10.7717\/peerj-cs.3793\/ref-8","doi-asserted-by":"publisher","first-page":"123721","DOI":"10.1016\/j.eswa.2024.123721","article-title":"LCDFormer: long-term correlations dual-graph transformer for traffic forecasting","volume":"249","author":"Cai","year":"2024","journal-title":"Expert Systems with Applications"},{"issue":"1","key":"10.7717\/peerj-cs.3793\/ref-9","doi-asserted-by":"publisher","first-page":"96","DOI":"10.3141\/1748-12","article-title":"Freeway performance measurement system: mining loop detector data","volume":"1748","author":"Chen","year":"2001","journal-title":"Transportation Research Record"},{"key":"10.7717\/peerj-cs.3793\/ref-10","first-page":"171","article-title":"ConvTimeNet: a deep hierarchical fully convolutional model for multivariate time series analysis","author":"Cheng","year":"2025"},{"key":"10.7717\/peerj-cs.3793\/ref-11","first-page":"894","article-title":"Soft-DTW: a differentiable loss function for time-series","author":"Cuturi","year":"2017"},{"key":"10.7717\/peerj-cs.3793\/ref-12","first-page":"11963","article-title":"Scaling up your kernels to 31x31: revisiting large kernel design in CNNs","author":"Ding","year":"2022"},{"key":"10.7717\/peerj-cs.3793\/ref-13","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2012.09699","article-title":"A generalization of transformer networks to graphs","author":"Dwivedi","year":"2020"},{"key":"10.7717\/peerj-cs.3793\/ref-14","first-page":"364","article-title":"Spatial-temporal graph ode networks for traffic flow forecasting","author":"Fang","year":"2021"},{"issue":"6","key":"10.7717\/peerj-cs.3793\/ref-15","doi-asserted-by":"publisher","first-page":"345","DOI":"10.1145\/367766.368168","article-title":"Algorithm 97: shortest path","volume":"5","author":"Floyd","year":"1962","journal-title":"Communications of the ACM"},{"key":"10.7717\/peerj-cs.3793\/ref-16","first-page":"3170","article-title":"SimVP: simpler yet better video prediction","author":"Gao","year":"2022"},{"issue":"11","key":"10.7717\/peerj-cs.3793\/ref-17","doi-asserted-by":"publisher","first-page":"5415","DOI":"10.1109\/tkde.2021.3056502","article-title":"Learning dynamics and heterogeneity of spatial-temporal graph data for traffic forecasting","volume":"34","author":"Guo","year":"2021","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"issue":"4","key":"10.7717\/peerj-cs.3793\/ref-18","doi-asserted-by":"publisher","first-page":"733","DOI":"10.1007\/s41095-023-0364-2","article-title":"Visual attention network","volume":"9","author":"Guo","year":"2023","journal-title":"Computational Visual Media"},{"issue":"8","key":"10.7717\/peerj-cs.3793\/ref-19","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Computation"},{"issue":"4","key":"10.7717\/peerj-cs.3793\/ref-20","doi-asserted-by":"publisher","first-page":"1700","DOI":"10.1109\/tits.2013.2267735","article-title":"Supervised weighting-online learning algorithm for short-term traffic flow prediction","volume":"14","author":"Jeong","year":"2013","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"key":"10.7717\/peerj-cs.3793\/ref-21","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v37i4.25556","article-title":"PDFormer: propagation delay-aware dynamic long-range transformer for traffic flow prediction","author":"Jiang","year":"2023a"},{"issue":"3","key":"10.7717\/peerj-cs.3793\/ref-22","doi-asserted-by":"publisher","first-page":"100","DOI":"10.3390\/ijgi12030100","article-title":"Graph neural network for traffic forecasting: the research progress","volume":"12","author":"Jiang","year":"2023b","journal-title":"ISPRS International Journal of Geo-Information"},{"key":"10.7717\/peerj-cs.3793\/ref-23","article-title":"Diffusion convolutional recurrent neural network: data-driven traffic forecasting","author":"Li","year":"2018"},{"issue":"9","key":"10.7717\/peerj-cs.3793\/ref-24","doi-asserted-by":"publisher","first-page":"9088","DOI":"10.1109\/tkde.2022.3221316","article-title":"DMGAN: dynamic multi-hop graph attention network for traffic forecasting","volume":"35","author":"Li","year":"2022","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"issue":"5","key":"10.7717\/peerj-cs.3793\/ref-25","doi-asserted-by":"publisher","first-page":"4189","DOI":"10.1609\/aaai.v35i5.16542","article-title":"Spatial-temporal fusion graph neural networks for traffic flow forecasting","volume":"35","author":"Li","year":"2021","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"10.7717\/peerj-cs.3793\/ref-26","first-page":"1","article-title":"ModernTCN: a modern pure convolution structure for general time series analysis","author":"Luo","year":"2024"},{"issue":"2","key":"10.7717\/peerj-cs.3793\/ref-27","doi-asserted-by":"publisher","first-page":"107074","DOI":"10.1016\/j.neunet.2024.107074","article-title":"ST-GMLP: a concise spatial-temporal framework based on gated multi-layer perceptron for traffic flow forecasting","volume":"184","author":"Luo","year":"2025","journal-title":"Neural Networks"},{"key":"10.7717\/peerj-cs.3793\/ref-28","article-title":"A time series is worth 64 words: long-term forecasting with transformers","author":"Nie","year":"2023"},{"issue":"1","key":"10.7717\/peerj-cs.3793\/ref-29","doi-asserted-by":"publisher","first-page":"1","DOI":"10.53375\/ijecer.2022.93","article-title":"Optimal electric vehicles route planning with traffic flow prediction and real-time traffic incidents","volume":"2","author":"Sebai","year":"2022","journal-title":"International Journal of Electrical and Computer Engineering Research"},{"issue":"01","key":"10.7717\/peerj-cs.3793\/ref-30","doi-asserted-by":"publisher","first-page":"914","DOI":"10.1609\/aaai.v34i01.5438","article-title":"Spatial-temporal synchronous graph convolutional networks: a new framework for spatial-temporal network data forecasting","volume":"34","author":"Song","year":"2020","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"issue":"12","key":"10.7717\/peerj-cs.3793\/ref-31","doi-asserted-by":"publisher","first-page":"23680","DOI":"10.1109\/tits.2022.3208943","article-title":"Dual dynamic spatial-temporal graph convolution network for traffic prediction","volume":"23","author":"Sun","year":"2022","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"key":"10.7717\/peerj-cs.3793\/ref-32","doi-asserted-by":"publisher","first-page":"5170","DOI":"10.1109\/tmm.2025.3543051","article-title":"SimVPv2: towards simple yet powerful spatiotemporal predictive learning","volume":"27","author":"Tan","year":"2025","journal-title":"IEEE Transactions on Multimedia"},{"issue":"4","key":"10.7717\/peerj-cs.3793\/ref-33","doi-asserted-by":"publisher","first-page":"113533","DOI":"10.1016\/j.knosys.2025.113533","article-title":"D-STANet: a delay-enhanced spatio-temporal attention network for traffic prediction","volume":"320","author":"Tang","year":"2025","journal-title":"Knowledge-Based Systems"},{"issue":"6","key":"10.7717\/peerj-cs.3793\/ref-34","doi-asserted-by":"publisher","first-page":"131765","DOI":"10.1016\/j.neucom.2025.131765","article-title":"A sparse dynamic graph transformer for traffic flow prediction","volume":"659","author":"Teng","year":"2025a","journal-title":"Neurocomputing"},{"issue":"20","key":"10.7717\/peerj-cs.3793\/ref-35","doi-asserted-by":"publisher","first-page":"11254","DOI":"10.3390\/app152011254","article-title":"Frequency-aware and interactive spatial-temporal graph convolutional network for traffic flow prediction","volume":"15","author":"Teng","year":"2025b","journal-title":"Applied Sciences"},{"issue":"10","key":"10.7717\/peerj-cs.3793\/ref-36","doi-asserted-by":"publisher","first-page":"2512","DOI":"10.14778\/3603581.3603591","article-title":"Effective and efficient route planning using historical trajectories on road networks","volume":"16","author":"Tian","year":"2023","journal-title":"Proceedings of the VLDB Endowment"},{"issue":"4","key":"10.7717\/peerj-cs.3793\/ref-37","doi-asserted-by":"publisher","first-page":"276","DOI":"10.1109\/tits.2004.837813","article-title":"Travel-time prediction with support vector regression","volume":"5","author":"Wu","year":"2004","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"key":"10.7717\/peerj-cs.3793\/ref-38","first-page":"753","article-title":"Connecting the dots: multivariate time series forecasting with graph neural networks","author":"Wu","year":"2020"},{"key":"10.7717\/peerj-cs.3793\/ref-39","first-page":"1907","article-title":"Graph WaveNet for deep spatial-temporal graph modeling","author":"Wu","year":"2019"},{"issue":"11","key":"10.7717\/peerj-cs.3793\/ref-40","doi-asserted-by":"publisher","first-page":"22386","DOI":"10.1109\/tits.2021.3102983","article-title":"Learning dynamic and hierarchical traffic spatiotemporal features with transformer","volume":"23","author":"Yan","year":"2021","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"issue":"8","key":"10.7717\/peerj-cs.3793\/ref-41","doi-asserted-by":"publisher","first-page":"8832","DOI":"10.1609\/aaai.v36i8.20864","article-title":"Graph pointer neural networks","volume":"36","author":"Yang","year":"2022","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"issue":"10","key":"10.7717\/peerj-cs.3793\/ref-42","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s11432-015-5397-4","article-title":"A literature survey on smart cities","volume":"58","author":"Yin","year":"2015","journal-title":"Science China Information Sciences"},{"issue":"3","key":"10.7717\/peerj-cs.3793\/ref-43","doi-asserted-by":"publisher","first-page":"e1913","DOI":"10.7717\/peerj-cs.1913","article-title":"Dynamic multiple-graph spatial-temporal synchronous aggregation framework for traffic prediction in intelligent transportation systems","volume":"10","author":"Yu","year":"2024","journal-title":"PeerJ Computer Science"},{"issue":"1","key":"10.7717\/peerj-cs.3793\/ref-44","doi-asserted-by":"publisher","first-page":"124187","DOI":"10.1016\/j.eswa.2024.124187","article-title":"Modeling dynamic spatio-temporal correlations and transitions with time window partitioning for traffic flow prediction","volume":"252","author":"Yu","year":"2024","journal-title":"Expert Systems with Applications"},{"issue":"9","key":"10.7717\/peerj-cs.3793\/ref-45","doi-asserted-by":"publisher","first-page":"13573","DOI":"10.1109\/tits.2025.3567841","article-title":"Distributed traffic signal control model for accurate policy learning under dynamic traffic flow: a graph forecast-state vector driven deep reinforcement learning framework","volume":"26","author":"Yu","year":"2025","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"key":"10.7717\/peerj-cs.3793\/ref-46","first-page":"3634","article-title":"Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting","author":"Yu","year":"2018"},{"issue":"9","key":"10.7717\/peerj-cs.3793\/ref-47","doi-asserted-by":"publisher","first-page":"11121","DOI":"10.1609\/aaai.v37i9.26317","article-title":"Are transformers effective for time series forecasting?","volume":"37","author":"Zeng","year":"2023","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"issue":"1\u20133","key":"10.7717\/peerj-cs.3793\/ref-48","doi-asserted-by":"publisher","first-page":"e1450","DOI":"10.7717\/peerj-cs.1450","article-title":"Spatial-temporal hypergraph convolutional network for traffic forecasting","volume":"9","author":"Zhao","year":"2023b","journal-title":"PeerJ Computer Science"},{"issue":"6","key":"10.7717\/peerj-cs.3793\/ref-49","doi-asserted-by":"publisher","first-page":"e1484","DOI":"10.7717\/peerj-cs.1484","article-title":"Spatio-temporal causal graph attention network for traffic flow prediction in intelligent transportation systems","volume":"9","author":"Zhao","year":"2023a","journal-title":"PeerJ Computer Science"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3793.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3793.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3793.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3793.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T08:33:47Z","timestamp":1776846827000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3793"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,22]]},"references-count":49,"alternative-id":["10.7717\/peerj-cs.3793"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3793","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,22]]},"article-number":"e3793"}}