Pith. sign in

REVIEW 4 major objections 5 minor 44 references

Forecasting at Full Spectrum: Holistic Multi-Granular Traffic Modeling under High-Throughput Inference Regimes

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Multi-scale traffic model forecasts 9.86% better, runs 2.43x faster

desk verdict A solid engineering combination with a weak theoretical centerpiece: the accuracy results are consistent and the distributed inference idea is worth attention, but the throughput guarantee is definitional and sloppy, and the lack of artifacts or error bars keeps the claims unverified. read the letter →

arxiv 2505.01279 v2 pith:56R6UISO submitted 2025-05-02 cs.LG

Zhaoyan Wang , Xiangchi Song , In-Young Ko This is my paper · ORCID
classification cs.LG
keywords trafficforecastinggraphconvolutionalnetworkmulti-granularfeaturefusiondynamicgenerationpipelineparallelismheterogeneousschedulinggeneticalgorithmintelligenttransportationsystems
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

MultiGran-STGCNFog claims to settle a trade-off: multi-granular spatiotemporal traffic models are more accurate but slow, and the paper answers by fusing features across scales inside a GCN while slicing the model across heterogeneous fog devices with a genetic pipeline scheduler. On PEMS04, PEMS07, and PEMS08, the proposed model reports up to 9.86% better MAE/MAPE/RMSE than the selected baselines, and the distributed inference system reports up to 2.43x throughput over GPipe/PipeDream-style scheduling. The accuracy gain comes from jointly modeling three spatial scales and three temporal scales with learnable gates; the throughput gain comes from optimizing both layer execution order and layer-device assignment, not just device assignment alone. The paper also derives a lower bound on the throughput gain in terms of output-tensor imbalance and link bandwidth asymmetry.

What carries the argument

The load-bearing object is GA-DPHDS, a two-level scheduler: an outer NSGA-II genetic search over layer execution orders constrained by the model's DAG, and an inner dynamic program that partitions the ordered layers into stages and maps stages to heterogeneous devices. The DP state $T[i][j]$ records the minimal possible maximum stage time for the first $i$ layers on the first $j$ devices, where each candidate stage time is the maximum of its compute time and the communication time $\mathrm{output}_k B_\mu / \min(b^{\mathrm{up}}_{j-1}, b^{\mathrm{down}}_j)$. This makes the scheduler choose splits that avoid placing large-output layers on slow links, which is exactly the condition the throughput theorem relies on. On the model side, the complementary machinery is the gated cross-granularity fusion block whose learnable gates $\mathbf{G} = \sigma(W_g X_{\mathrm{concat}} + b_g)$ combine spatial scales extracted by Laplacian-eigenvector clustering and temporal scales extracted by hourly, daily, and weekly windows.

What would settle it

Run GPipe and PipeDream on the same six-device clusters with the same profiled per-layer compute times and per-link bandwidths, and record which layer boundary actually forms the bottleneck. If the slowest stage in the baseline is not a large-output layer on the slowest link, the claimed lower bound $\gamma_D \ge \alpha_D \beta / (1+\varepsilon)$ would not describe the real comparison; likewise, measuring the actual post-scheduling imbalance $\delta$ and recomputing the bound would settle whether the guarantee survives outside the assumed worst case.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that a traffic forecaster built on dynamic graph convolution can be made both more accurate and faster to serve by treating feature fusion and layer scheduling as one design problem. MultiGran-STGCN constructs a dynamic adjacency matrix at every time step, forms coarse spatial nodes by clustering eigenvectors of the normalized graph Laplacian, and slices history into hourly, daily, and weekly windows; cross-granularity gated blocks then fuse these views before a weighted forecasting head. For serving, GA-DPHDS uses NSGA-II to evolve candidate layer execution orders, and a dynamic program $T[i][j]$ assigns the first $i$ layers to $j$ devices while minimizing the maximum of compute time and communication time across pipeline stages. Experiments on the three PEMS datasets show the full model ahead of all listed baselines at 15-, 30-, and 60-minute horizons, with the 60-minute PEMS08 MAE improving 9.86% over the top-3 baselines. The throughput experiments across ten physical clusters show GA-DPHDS ahead of GPipe/PipeDream in every cluster, and the paper derives a theoretical lower bound $\gamma_D \ge \alpha_D \beta / (1+\varepsilon)$ on the gain.

Load-bearing premise

The throughput lower bound assumes the baseline scheduler places a large-output layer on the slowest communication link, while GA-DPHDS, by construction, avoids such splits; the residual imbalance $\delta$ after scheduling is treated as a small known constant rather than measured from the real profiled device times.

Editorial extensions

If this is right

  • If the accuracy results hold, multi-granular fusion over three spatial and three temporal scales is a reliable way to improve GCN traffic forecasting, with larger gains at longer horizons.
  • If the throughput results hold, layer execution order is a first-class decision in pipeline-parallel inference, not a detail: optimizing it raised cluster throughput by up to 41% in the ablation.
  • On bandwidth-starved clusters the scheduler collapses to single-device execution, so the throughput benefit is conditional on having enough link bandwidth to make pipeline parallelism worthwhile.
  • The lower bound implies throughput gains grow with output-tensor imbalance across layers and with asymmetry between slow and fast links, so the method should matter most for models with heterogeneous layer sizes deployed on uneven clusters.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension is to apply the same scheduler to other GNN architectures with different output-tensor profiles; the theorem predicts larger gains for models whose layers have high output-size imbalance.
  • A natural next question is whether the fusion mechanism transfers to urban networks, extreme-event periods, or missing-data regimes, none of which the three PEMS highway datasets cover.
  • The scheduler is evaluated for inference; extending it to training would require accounting for backward-pass memory and gradient communication, which the current objective does not model.
  • A direct measurement of per-stage times on the physical cluster would show whether the worst-case baseline used in the lower-bound proof is the right comparison point for real GPipe and PipeDream placements.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes MultiGran-STGCNFog, a system that combines a new GCN-based traffic forecasting model with a genetic-algorithm scheduler for pipeline-parallel inference on heterogeneous fog devices. The model, MultiGran-STGCN, extracts spatial features at multiple graph granularities via Laplacian-based clustering and temporal features at hourly, daily, and weekly scales, fusing them with learnable gates. The scheduler, GA-DPHDS, jointly optimizes layer execution order and layer-to-device assignment using NSGA-II and dynamic programming. The authors report accuracy improvements over several GCN baselines on PEMS04, PEMS07, and PEMS08 (up to 9.86% in MAE), and throughput gains over GPipe/PipeDream-style baselines on a six-device fog testbed (up to 2.43x claimed). Section 4.4 derives a theoretical lower bound on throughput improvement, giving a numerical estimate of 1.54x. The paper also includes ablation studies for the multi-granular design and for the layer-order optimization.

Significance. If validated, the paper would make a useful contribution by combining multi-granular spatiotemporal modeling with a practical distributed inference scheduler, an integration that is rarely addressed in the traffic forecasting literature. The use of real heterogeneous devices, ten cluster configurations, and ablation studies for both model components and scheduling components are strengths. The theoretical throughput analysis attempts to provide a formal guarantee, which is commendable, and the reported accuracy gains, while incremental, are consistent across three datasets and three horizons. However, the current evidence is weakened by the definitional nature of the throughput 'bound,' the mismatch between the theoretical baseline and the implemented baselines, and the absence of error bars or significance tests. These issues do not invalidate the empirical results out of hand, but they do require substantial revision before the central claims can be accepted.

major comments (4)
  1. [Section 4.4, Eqs. (35)-(38)] The claimed theoretical lower bound is largely a restatement of the definition of the throughput improvement ratio in Eq. (32), dressed up with assumed quantities. Eq. (35) follows algebraically once one assumes the baseline bottleneck in Eq. (33) and the GA upper bound in Eq. (34). The numerical estimate in Eq. (38) is not measured: δ is assumed to be 10 ms but is written as 0.005 s, an internal inconsistency, and the term o_D^avg / min(b_good) = 0.00555 s omits the micro-batch factor B_mu = 16 that is present in the communication model of Eqs. (6) and (29). Including that factor would yield approximately 0.0888 s, substantially changing the denominator and the resulting bound. The authors should measure δ from the actual GA-DPHDS schedules, correct the units, include B_mu, and then recompute the estimate before presenting 1.54x as a validated guarantee.
  2. [Section 5.3.1 and Section 4.4] The theoretical comparison in Section 4.4 is against a baseline whose worst-case bottleneck is a high-output layer placed on a slow link, as assumed in Eq. (33). The experimental baselines, however, are described only as 'the scheduling methods of GPipe and PipeDream adopted as baseline' with 'devices and pipeline sequences randomized.' Randomized partitions need not realize the worst case in Eq. (33), and GPipe and PipeDream in practice use balanced or profile-based partitioning rather than random equal splits. Consequently, the empirical throughput ratios (e.g., 67.14 vs 30.68 samples/s in Cluster 4) may be measured against a strawman baseline, and the claimed 2.43x improvement is not clearly attributable to any reported cluster. The manuscript must specify the exact partitioning and device-assignment rules used for the GPipe and PipeDream baselines and verify whether those baselines satisfy the assumptions of Eq. (33).
  3. [Section 4.4, Eq. (30)] The communication-time model in Eq. (30) uses only the output size of the final layer in a stage, o_D_{j_last}. This is inadequate for MultiGran-STGCN, whose architecture contains three temporal branches feeding a fusion head (Section 4.2.3, Fig. 2). A stage produced by GA-DPHDS can contain layers from multiple branches, and a layer in the DAG can have output tensors flowing to several successor layers that reside on a different device. In such cases, the total communication volume across the device boundary exceeds the output size of the single last-ordered layer. The analysis should either restrict stages to linear chains or sum the sizes of all output tensors that cross the device boundary.
  4. [Section 5, Tables 2 and Fig. 5] The empirical results are reported as single-point averages without error bars or statistical tests. This is particularly concerning for the forecasting comparisons, where some margins are very small (e.g., PEMS04 15-min MAE of 18.11 vs 18.12 for GWNET in Table 2), and for the throughput experiments, which are averaged over ten runs but reported without variance. The central claims of accuracy gains and throughput gains would be substantially strengthened by reporting standard deviations across multiple seeds or runs and, where appropriate, paired significance tests.
minor comments (5)
  1. [Section 4.4, Eq. (38)] The text states 'Assume δ = 10ms = 0.005,' but 10 ms is 0.01 s; the value 0.005 s is 5 ms and is internally inconsistent.
  2. [Abstract and Section 5.3.1] The abstract claims '2.43x throughput improvement,' but the reported Cluster 4 result is 67.14 vs 30.68 samples/s, which is approximately 2.19x; the paper should identify which cluster configuration produces the 2.43x figure or correct the claim.
  3. [Table 3] The cluster configurations in Table 3 are nearly illegible because they use checkmarks without explicit device lists; a table listing the exact device IDs per cluster would be clearer.
  4. [Section 5.3.2] The statement that 'in Clusters 1–4, all layers were assigned to device A' at low bandwidth appears to contradict the notion of pipelined inference; the authors should explain how this assignment still counts as pipeline parallelism or clarify that the system degenerates to single-device execution.
  5. [Section 5.1] The paper does not report model hyperparameters such as window lengths, number of clusters, hidden dimensions, and training details; these should be provided in an appendix for reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: forecasting gains are empirical and the throughput bound is a derived inequality from stated assumptions, not a refit of the measured outcome.

full rationale

I walked the paper's claimed derivation chains. The forecasting contribution (Sections 4.2 and 5.2) is an empirical architecture comparison: MultiGran-STGCN is trained and evaluated against external baselines on PEMS04/07/08, and the reported MAE/MAPE/RMSE improvements are measured outcomes, not quantities defined in terms of the model's own outputs. The scheduling contribution (Section 4.3) profiles per-layer times and uses GA-DPHDS to optimize (O*, S*); this is a normal optimization procedure, not a circular prediction. The theoretical throughput analysis in Section 4.4 defines gamma_D in Eq. (32), states a worst-case baseline lower bound in Eq. (33), a GA-DPHDS upper bound in Eq. (34), and then derives Eq. (35) by substituting those bounds into the definition. This is algebra from explicit assumptions, so it is not self-definitional. The numerical estimate in Eq. (38) plugs in measured output sizes and assumed delta; that makes the 1.54x figure conditional, but it is not a fitted parameter masquerading as a prediction. The paper contains no load-bearing self-citations: all cited works are independent prior literature, and no uniqueness theorem or prior result by the same authors is invoked to forbid alternatives. The weaknesses noted by the reader — delta is assumed rather than measured, the baseline in Eq. (33) may not match the randomized GPipe/PipeDream implementation, and Eq. (38) appears to omit the micro-batch factor B_mu in the denominator — are correctness and robustness concerns, not circularity. Therefore the derivation chain is self-contained and no circular step is exhibited.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central claims rest on a set of hand-chosen model hyperparameters, measured profiling values, and an assumed worst-case baseline for the throughput theorem. The model itself introduces no physical or theoretical entities beyond standard GCN mechanisms.

free parameters (5)
  • output tensor imbalance ratio alpha_D = 3.24 (from authors' PEMS04 run)
    Computed from output tensor sizes of the authors' model; used in Eq. (35) to obtain the throughput bound. It is a measured quantity from the same experiments, not an independent constant.
  • link bandwidth asymmetry ratio beta = 1.33
    Derived from assumed good/bad link bandwidths (200 MB/s vs 150 MB/s) in Section 4.4; used to estimate the improvement bound.
  • residual imbalance delta = 10 ms (assumed)
    Assumed in Section 4.4 to estimate the lower bound; no measurement is provided.
  • micro-batch size B_mu = 16
    Set in the experimental setup; directly affects communication time in Eqs. (6), (29), (34) and the theorem.
  • model hyperparameters (window lengths, cluster counts, hidden dims, embedding dims, gating dims) = not reported
    The architecture in Fig. 2 is described qualitatively; these hand-chosen values are not listed, so the accuracy results depend on undisclosed choices.
assumptions (5)
  • domain assumption Laplacian eigenvector clustering produces semantically meaningful spatial groupings for traffic networks.
    Assumed in Section 4.2.1; used to construct coarser spatial scales. If clusters are meaningless, the multi-spatial branches add noise.
  • domain assumption Traffic exhibits stable daily and weekly periodicity, so hourly/daily/weekly windows capture the relevant temporal scales.
    Invoked in Section 4.2.1 to define short, medium, and long-term inputs.
  • domain assumption The dynamic adjacency matrix from node embedding inner products (Eq. 14) captures time-varying spatial dependencies.
    Used in Section 4.2.2 for graph convolution; if the embeddings do not encode traffic relations, the adaptive graph is not useful.
  • domain assumption Per-layer profiling times and bandwidth measurements are stable and representative of runtime conditions.
    The scheduling in Section 4.1 relies on these measurements to compute T_comp and T_comm; runtime variation would invalidate the optimality of the schedule.
  • ad hoc to paper The baseline scheduler's worst-case stage time is dominated by a high-output layer on a slow link (Eq. 33), and GA-DPHDS avoids such splits.
    This is the key assumption in the Theorem 4.1 proof; it is not verified against the actual baseline implementations used in the experiments.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Forecasting at Full Spectrum: Holistic Multi-Granular Traffic Modeling under High-Throughput Inference Regimes." pith.science (2026). https://pith.science/paper/56R6UISO

@misc{pith2026250501279,
  author       = {Pith},
  title        = {Pith review of: Forecasting at Full Spectrum: Holistic Multi-Granular Traffic Modeling under High-Throughput Inference Regimes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/56R6UISO}},
  note         = {Machine review of arXiv:2505.01279}
}
read the original abstract

Notably, current intelligent transportation systems rely heavily on accurate traffic forecasting and swift inference provision to make timely decisions. While Graph Convolutional Networks (GCNs) have shown benefits in modeling complex traffic dependencies, the existing GCN-based approaches cannot fully extract and fuse multi-granular spatiotemporal features across various spatial and temporal scales sufficiently in a complete manner, proven to yield less accurate results. Besides, as extracting multi-granular features across scales has been a promising strategy across domains such as computer vision, natural language processing, and time-series forecasting, pioneering studies have attempted to leverage a similar mechanism for spatiotemporal traffic data mining. However, additional feature extraction branches introduced in prior studies critically increased model complexity and extended inference time, making it challenging to provide fast forecasts. In this paper, we propose MultiGran-STGCNFog, an efficient fog distributed inference system with a novel traffic forecasting model that employs multi-granular spatiotemporal feature fusion on generated dynamic traffic graphs to fully capture interdependent traffic dynamics. The proposed scheduling algorithm GA-DPHDS, optimizing layer execution order and layer-device scheduling scheme simultaneously, contributes to considerable inference throughput improvement by coordinating heterogeneous fog devices in a pipelined manner. Extensive experiments on real-world datasets demonstrate the superiority of the proposed method over selected GCN baselines.

Figures

Figures reproduced from arXiv: 2505.01279 by the authors.

Figure 1
Figure 1. The overview of MultiGran-STGCNFog. 3.2.2 Load Balancing. To mitigate pipeline inefficiencies caused by load imbalance, which introduces idle times on faster devices, we aim to balance the load across devices by minimizing the standard deviation 𝜎 of devices’ execution time 𝑇exec (𝑑𝑖): 𝜎 = vut 1 𝑁𝑑 ∑︁ 𝑁𝑑 𝑖=1  𝑇exec (𝑑𝑖) −𝑇exec2 . (7) 3.2.3 Optimization Summary. In conclusion, to maximize the pipeline throughput, w… view at source ↗
Figure 2
Figure 2. The network architecture of MultiGran-STGCN. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. The illustrative diagram of inference acceleration [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Ablation study: forecasts across horizons. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 6
Figure 6. Figure 6: Cluster throughput with bandwidth scaling factors. [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Ablation study: layer execution order optimization. [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 21 canonical work pages

  1. [1]

    Abien Fred M Agarap. 2018. A neural network architecture combining gated recurrent unit (GRU) and support vector machine (SVM) for intrusion detection in network traffic data. In Proceedings of the 2018 10th international conference on machine learning and computing . 26–30

    work page 2018

  2. [2]

    Iz Beltagy, Matthew E Peters, and Arman Cohan. 2020. Longformer: The long- document transformer. arXiv preprint arXiv:2004.05150 (2020)

    arXiv 2020

  3. [3]

    Maciej Besta and Torsten Hoefler. 2024. Parallel and distributed graph neural net- works: An in-depth concurrency analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence (2024)

    work page 2024

  4. [4]

    Jingji Chen, Zhuoming Chen, and Xuehai Qian. 2023. GNNPipe: Scaling Deep GNN Training with Pipelined Model Parallelism. arXiv preprint arXiv:2308.10087 (2023)

    arXiv 2023

  5. [5]

    Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. 2018. Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European conference on computer vision (ECCV). 801–818

    2018

  6. [6]

    Minxiao Chen, Haitao Yuan, Nan Jiang, Zhifeng Bao, and Shangguang Wang

    work page

  7. [7]

    Zhicheng Cui, Wenlin Chen, and Yixin Chen. 2016. Multi-scale convolutional neural networks for time series classification. arXiv preprint arXiv:1603.06995 (2016)

    arXiv 2016

  8. [8]

    Zhiyong Cui, Kristian Henrickson, Ruimin Ke, and Yinhai Wang. 2020. Traffic Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting. IEEE Transactions on Intelligent Transportation Systems 21, 11 (2020), 4883–4894. doi:10.1109/TITS.2019.2950416

    arXiv 2020

Show all 44 references
  1. [9]

    Zhiyong Cui, Ruimin Ke, Ziyuan Pu, and Yinhai Wang. 2018. Deep bidirectional and unidirectional LSTM recurrent neural network for network-wide traffic speed prediction. arXiv preprint arXiv:1801.02143 (2018)

    2018 arXiv

  2. [10]

    Zihang Dai, Zhilin Yang, Yiming Yang, Jaime Carbonell, Quoc V Le, and Ruslan Salakhutdinov. 2019. Transformer-xl: Attentive language models beyond a fixed- length context. arXiv preprint arXiv:1901.02860 (2019)

    2019 arXiv

  3. [11]

    Tangpeng Dan, Xiao Pan, Bolong Zheng, and Xiaofeng Meng. 2024. ByGCN: Spatial Temporal Byroad-Aware Graph Convolution Network for Traffic Flow Pre- diction in Road Networks. InProceedings of the 33rd ACM International Conference on Information and Knowledge Management . 415–424

    2024

  4. [12]

    Kan Guo, Yongli Hu, Zhen Qian, Hao Liu, Ke Zhang, Yanfeng Sun, Junbin Gao, and Baocai Yin. 2020. Optimized graph convolution recurrent neural network for traffic prediction. IEEE Transactions on Intelligent Transportation Systems 22, 2 (2020), 1138–1149

    2020

  5. [13]

    Kan Guo, Yongli Hu, Yanfeng Sun, Sean Qian, Junbin Gao, and Baocai Yin. 2021. Hierarchical graph convolution network for traffic forecasting. In Proceedings of the AAAI conference on artificial intelligence , Vol. 35. 151–159

    2021

  6. [14]

    Shengnan Guo, Youfang Lin, Ning Feng, Chao Song, and Huaiyu Wan. 2019. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In Proceedings of the AAAI conference on artificial intelligence , Vol. 33. 922–929

    2019

  7. [15]

    Yanping Huang, Youlong Cheng, Ankur Bapna, Orhan Firat, Dehao Chen, Mia Chen, HyoukJoong Lee, Jiquan Ngiam, Quoc V Le, Yonghui Wu, et al. 2019. Gpipe: Efficient training of giant neural networks using pipeline parallelism. Advances in neural information processing systems 32 (2019)

    2019

  8. [16]

    Guangyu Huo, Yong Zhang, Boyue Wang, Junbin Gao, Yongli Hu, and Baocai Yin. 2023. Hierarchical spatio–temporal graph convolutional networks and transformer network for traffic flow forecasting. IEEE Transactions on Intelligent Transportation Systems 24, 4 (2023), 3855–3867

    2023

  9. [17]

    Robin Jia, Cliff Wong, and Hoifung Poon. 2019. Document-level𝑁 -ary relation ex- traction with multiscale representation learning. arXiv preprint arXiv:1904.02347 (2019)

    2019 arXiv

  10. [18]

    Jiawei Jiang, Chengkai Han, Wayne Xin Zhao, and Jingyuan Wang. 2023. Pdformer: Propagation delay-aware dynamic long-range transformer for traffic flow prediction. In Proceedings of the AAAI conference on artificial intelligence , Vol. 37. 4365–4373

    2023

  11. [19]

    Ning Jin, Jiaxian Wu, Xiang Ma, Ke Yan, and Yuchang Mo. 2020. Multi-task learning model based on multi-scale CNN and LSTM for sentiment classification. IEEE Access 8 (2020), 77060–77072

    2020

  12. [20]

    Fuxian Li, Huan Yan, Guangyin Jin, Yue Liu, Yong Li, and Depeng Jin. 2022. Automated spatio-temporal synchronous modeling with multiple graphs for traffic prediction. In Proceedings of the 31st ACM International Conference on Information & Knowledge Management . 1084–1093

    2022

  13. [21]

    Shuhao Li, Yue Cui, Jingyi Xu, Jing Zhao, Fan Zhang, Weidong Yang, and Xiaofang Zhou. 2024. Seeing the Forest for the Trees: Road-Level Insights Assisted Lane- Level Traffic Prediction. In Proceedings of the 33rd ACM International Conference on Information and Knowledge Manage...

    2024

  14. [22]

    Yaguang Li, Rose Yu, Cyrus Shahabi, and Yan Liu. 2017. Diffusion convolu- tional recurrent neural network: Data-driven traffic forecasting. arXiv preprint arXiv:1707.01926 (2017)

    2017 arXiv

  15. [23]

    Bryan Lim, Sercan Ö Arık, Nicolas Loeff, and Tomas Pfister. 2021. Temporal fusion transformers for interpretable multi-horizon time series forecasting.International Journal of Forecasting 37, 4 (2021), 1748–1764

    2021

  16. [24]

    Tsung-Yi Lin, Piotr Dollar, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie. 2017. Feature Pyramid Networks for Object Detection. InProceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    2017

  17. [25]

    Hangchen Liu, Zheng Dong, Renhe Jiang, Jiewen Deng, Jinliang Deng, Quan- jun Chen, and Xuan Song. 2023. Spatio-temporal adaptive embedding makes vanilla transformer sota for traffic forecasting. In Proceedings of the 32nd ACM international conference on information and knowled...

    2023

  18. [26]

    Deepak Narayanan, Aaron Harlap, Amar Phanishayee, Vivek Seshadri, Nikhil R Devanur, Gregory R Ganger, Phillip B Gibbons, and Matei Zaharia. 2019. PipeDream: Generalized pipeline parallelism for DNN training. In Proceedings of the 27th ACM symposium on operating systems princip...

    2019

  19. [27]

    Youngjoo Seo, Michaël Defferrard, Pierre Vandergheynst, and Xavier Bresson

  20. [28]

    Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke, and Alexander Alemi. 2017. Inception-v4, inception-resnet and the impact of residual connections on learning. In Proceedings of the AAAI conference on artificial intelligence , Vol. 31

    2017

  21. [29]

    Senzhang Wang, Meiyue Zhang, Hao Miao, and Philip S Yu. 2021. Mt-stnets: Multi- task spatial-temporal networks for multi-scale traffic prediction. In Proceedings of the 2021 SIAM International Conference on Data Mining (SDM) . SIAM, 504–512

    2021

  22. [30]

    Yi Wang and Changfeng Jing. 2022. Spatiotemporal graph convolutional network for multi-scale traffic forecasting. ISPRS International Journal of Geo-Information 11, 2 (2022), 102

    2022

  23. [31]

    Haixu Wu, Tengge Hu, Yong Liu, Hang Zhou, Jianmin Wang, and Mingsheng Long. 2022. Timesnet: Temporal 2d-variation modeling for general time series analysis. arXiv preprint arXiv:2210.02186 (2022)

    2022 arXiv

  24. [32]

    Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang, and Chengqi Zhang. 2019. Graph wavenet for deep spatial-temporal graph modeling. arXiv preprint arXiv:1906.00121 (2019)

    2019 arXiv

  25. [33]

    Bing Yu, Haoteng Yin, and Zhanxing Zhu. 2017. Spatio-temporal graph con- volutional networks: A deep learning framework for traffic forecasting. arXiv preprint arXiv:1709.04875 (2017)

    2017 arXiv

  26. [34]

    Fotios Zantalis, Grigorios Koulouras, Sotiris Karabetsos, and Dionisis Kandris

  27. [35]

    Liekang Zeng, Peng Huang, Ke Luo, Xiaoxi Zhang, Zhi Zhou, and Xu Chen

  28. [36]

    Weijia Zhang, Le Zhang, Jindong Han, Hao Liu, Yanjie Fu, Jingbo Zhou, Yu Mei, and Hui Xiong. 2024. Irregular Traffic Time Series Forecasting Based on Asynchronous Spatio-Temporal Graph Convolutional Networks. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discov...

    2024

  29. [37]

    Zhengyan Zhang, Xu Han, Zhiyuan Liu, Xin Jiang, Maosong Sun, and Qun Liu

  30. [38]

    Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, and Jiaya Jia. 2017. Pyramid scene parsing network. InProceedings of the IEEE conference on computer vision and pattern recognition . 2881–2890

    2017

  31. [39]

    Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wancai Zhang. 2021. Informer: Beyond efficient transformer for long se- quence time-series forecasting. In Proceedings of the AAAI conference on artificial intelligence, Vol. 35. 11106–11115. 10

    2021

  32. [42]

    arXiv preprint arXiv:1905.07129 (2019)

    ERNIE: Enhanced language representation with informative entities. arXiv preprint arXiv:1905.07129 (2019)

    2019 arXiv

  33. [2018]

    In Neural Information Processing: 25th International Conference, ICONIP 2018, Siem Reap, Cambodia, December 13-16, 2018, Proceedings, Part I 25

    Structured sequence modeling with graph convolutional recurrent net- works. In Neural Information Processing: 25th International Conference, ICONIP 2018, Siem Reap, Cambodia, December 13-16, 2018, Proceedings, Part I 25 . Springer, 362–373

    2018

  34. [2019]

    Future Internet 11, 4 (2019), 94

    A review of machine learning and IoT in smart transportation. Future Internet 11, 4 (2019), 94

    2019

  35. [2022]

    In Proceedings of the ACM Web Conference 2022

    Fograph: Enabling real-time deep graph inference with fog computing. In Proceedings of the ACM Web Conference 2022 . 1774–1784

    2022

  36. [2024]

    In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management

    Urban Traffic Accident Risk Prediction Revisited: Regionality, Proximity, Similarity and Sparsity. In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management . 281–290

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.