Pith. sign in

REVIEW 3 major objections 7 minor 35 references

SafeDivertor: Faithful Divertor Heat Flux Reconstruction from Macroscopic Plasma State Signals via Time-Frequency Prior Exploitation

T0 review · 3 major / 7 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read Divertor heat flux can be reconstructed during a discharge from routine plasma-state signals, without infrared imaging or heat-conduction inversion; the authors' framework beats eight time-series baselines on all five metrics.

desk verdict Worth a referee: real new dataset and task, but the benchmark claim is undercut by the absence of a trivial prior-only baseline and by heavily overlapping test windows. read the letter →

arxiv 2608.05669 v1 pith:W73DZMHO submitted 2026-08-06 physics.plasm-ph cs.AIcs.CV

Hao Si , Zehua Chen , Qingquan Yang , Xiao Wang , Dengdi Sun , Wanli Lyu , Gaoting Chen , Guosheng Xu
show 3 more authors
Hang Su Jin Tang Jun Zhu
This is my paper · ORCID
classification physics.plasm-phcs.AIcs.CV PACS 52.55.Fa52.40.Hf
keywords divertorheatfluxsignal-basedreconstructionmultivariatetimeseriesphysicalpriorshort-timeFouriertransformspectrallosstokamakplasmadiagnosticsonlinemonitoring
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

This paper claims that the time-resolved radial heat-flux profile on a tokamak divertor can be reconstructed while the discharge is running, directly from macroscopic plasma-state signals that are already recorded, with no infrared camera and no heat-conduction solver. To support the claim, the authors assemble DivMPS2HF, a dataset of 77 discharges that aligns 67 input channels with 116-channel radial heat-flux labels, and propose SafeDivertor, which fills zero-valued target placeholders with a training-set mean heat-flux prior, perturbs input channels during training, and supervises reconstruction with a multi-scale short-time Fourier transform loss. On held-out shots, SafeDivertor beats the best of eight time-series baselines on all five metrics, reducing MSE from 0.262 to 0.200 and high-frequency log-spectral distance from 3.829 to 2.729. If the claim holds, divertor heat-load monitoring and protection become an online inference problem rather than a post-shot thermal-analysis task.

What carries the argument

The load-bearing mechanism is the combination of physical prior-aware initialization with spectral-aware reconstruction optimization. The prior is a per-channel mean heat-flux profile computed from the training set, whose encoded feature is added only to the target-placeholder representations, giving the model radial-distribution guidance without touching the observed channels. The spectral loss matches the log-magnitude short-time Fourier transform spectra of reconstruction and ground truth at three temporal scales, which counteracts the smoothing induced by point-wise MSE and preserves transient high-frequency dynamics. Input perturbation regularizes the model against over-reliance on any particular diagnostic channel, while progressive training introduces MSE first, perturbation second, and the spectral loss third, so that the complementary objectives do not destabilize each other.

What would settle it

Run SafeDivertor with and without the physical prior on a held-out set of shots whose mean radial heat-flux profile differs clearly from the training distribution, for example a different confinement regime or a different strike-point configuration; the central claim would be undermined if the prior-equipped version performs worse than the no-prior baseline on that shifted set while still winning on in-distribution shots.

Watch

Extended reading notes

Core claim

The central discovery, on the paper's own terms, is that divertor heat-flux reconstruction can be reformulated as a structured channel-level reconstruction problem: concatenate the observed plasma-state signals with zero-valued placeholders for the 116 heat-flux channels, then train a single multivariate time-series model to fill in the placeholders. The authors report that SafeDivertor does this faithfully, preserving both the overall spatiotemporal structure of the heat-flux pattern and its high-frequency transient variations, and that it establishes a new benchmark with MSE 0.200, MAE 0.291, SSIM 0.870, LSD 2.475, and LSD-HF 2.729, the best result among the nine compared models on every metric.

Load-bearing premise

The load-bearing premise is that the training-set mean radial heat-flux profile, used as the physical prior, is representative of the test shots; if test conditions shift the mean heat-load shape, the prior will bias the reconstruction instead of guiding it.

Editorial extensions

If this is right

  • Heat-flux profiles become available in a single neural forward pass at 11.313 ms per 0.5-second window, a latency compatible with window-level online monitoring during a discharge.
  • The multi-scale spectral loss is the main lever for transient fidelity: adding it reduces LSD from 3.521 to 2.306 and LSD-HF from 3.829 to 2.511.
  • The training-set mean heat-flux prior alone lowers MSE from 0.262 to 0.211 and raises SSIM from 0.848 to 0.870, showing that statistical radial guidance carries much of the pointwise gain.
  • Off-the-shelf forecasting and exogenous-variable models are insufficient for this task; the strongest baseline is a time-series imputation model, and SafeDivertor improves on it across all five metrics.
  • DivMPS2HF gives the community a fixed shot-level split and a five-metric protocol for comparing future signal-based divertor heat-flux reconstruction methods.

Reading between the lines

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

  • A consequence the paper leaves implicit is that the same signal-to-flux mapping could serve as a cross-check or fallback for infrared inversion whenever IR diagnostics are unavailable, saturated, or being serviced.
  • A testable extension is regime-adaptive priors: since the ablation shows the training-set mean profile drives the pointwise improvement, updating the prior per confinement regime (attached versus detached) or per discharge could be the clearest next gain in accuracy.
  • The structured-reconstruction scaffolding may transfer to other plasma quantities currently available only post-shot, such as radiation emissivity or particle-flux profiles, giving operators online estimates of quantities they now reconstruct offline.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The paper introduces DivMPS2HF, a multi-source discharge dataset for a new task: reconstructing time-resolved radial divertor heat-flux profiles (116 channels) directly from 67 macroscopic plasma-state signal channels over 0.5 s windows, without infrared thermography as input. On this dataset, the authors propose SafeDivertor, a T1 imputation backbone augmented by physical prior-aware initialization (PPI), input perturbation (IP), multi-scale STFT-based spectral loss (SRO), and progressive training (PT). The paper reports that SafeDivertor outperforms eight time-series baselines across all five metrics (MSE, MAE, SSIM, LSD, LSD-HF) and provides ablations, efficiency analysis, and qualitative visualizations. The code is promised on GitHub; the dataset is said to be available upon reasonable request.

Significance. If the claimed signal-to-flux mapping is genuinely learned, the paper makes a useful contribution: it formulates a new online-oriented reconstruction paradigm, constructs a shot-split benchmark dataset, compares against eight recent time-series models, and shows that the proposed components yield consistent quantitative and qualitative improvements over the T1 backbone. The efficiency analysis is also a strength, as SafeDivertor adds negligible overhead over T1. However, the central claim is currently conditional. The largest ablation gain comes from injecting the per-channel training-set mean of the target labels (PPI), and the paper does not benchmark the trivial predictor that always outputs that mean. Until that baseline and shot-level statistics are provided, the headline claim that the model learns a signal-driven mapping is not established.

major comments (3)
  1. [§V-D, Eq. (4)–(7), Table II] The ablation evidence is insufficient to support the central claim that SafeDivertor learns a signal-to-flux mapping. PPI alone reduces the T1 backbone MSE from 0.262 to 0.211, a 19% drop, and PPI injects g_theta(P), where P is constructed from the per-channel training-set mean of the target heat-flux labels (Eqs. (4)–(5)). Because divertor heat-flux profiles are strongly dominated by a roughly fixed radial shape, a constant predictor that outputs the training-set mean radial profile for every test window may itself achieve an MSE close to 0.211 or even lower. Table I contains no such constant-prior baseline. Please add this trivial baseline with all five metrics, and additionally ablate the signal channels (e.g., feed noise or zeros in place of the observed X while retaining PPI) to demonstrate that the macroscopic signals contribute beyond the injected label statistic. The term 'physical prior' is also misleading: Eq. (4) is a statistical average of training labels, not a physics-based prior, and should be renamed or explicitly characterized as a label-statistic prior.
  2. [§V-A, §V-C] The evaluation is pseudoreplicated. The 28,982 test windows come from only 10 shots and are generated with a stride of one time step, so adjacent windows overlap in 499 of 500 time steps and are not independent. All metrics in Table I and Table II are pooled over these non-independent windows without shot-level aggregation or confidence intervals. With only 10 test shots, a model can obtain a spuriously good pooled score by performing well on a few shots. Please report per-shot results, shot-level means and standard errors, and state how many of the 10 test shots each improvement holds for.
  3. [§V-C, §V-D, Tables I–II] No error bars or repeated-seed results are reported. Several headline differences are small in absolute terms — for example, the full model's SSIM of 0.870 equals the PPI-only ablation's SSIM of 0.870 in Table II, and the MSE difference between the full model and PPI-only is 0.200 vs. 0.211. With a single training run one cannot distinguish genuine gains from optimization noise. Please run at least three random seeds and report mean ± standard deviation for the main comparison and the ablation tables.
minor comments (7)
  1. [Eq. (11)] The multi-scale STFT notation Y_r and \hat Y_r is not defined; please specify the STFT window sizes, hop lengths, and how the three temporal scales are constructed.
  2. [§IV-A] The paper states that the 67 input channels were 'selected from a broader pool of candidate signals' but does not report the size of the candidate pool or the selection criterion; this information is needed for reproducibility of the dataset.
  3. [Figures 3 and 4] The qualitative samples are described as 'randomly selected'; please state the random seed or selection procedure so that the figures can be reproduced.
  4. [§V-A and §VI] The dataset is described as a benchmark but is to be made available only 'upon reasonable request', while the code is on GitHub; please clarify the data-release conditions in the final version, since a benchmark paper benefits from a public dataset repository.
  5. [§V-G, Table V] The latency of 11.313 ms is compared only against other models, not against the 0.5 s window duration as a real-time budget; please state the end-to-end latency (including preprocessing) and its margin relative to the online requirement.
  6. [Table IV] The caption introduces 'CM' and 'GP' but the table body would be clearer if the column headers repeated the labels; as printed, the reader must map the caption abbreviations to the two sub-columns under each ratio.
  7. [§V-D, Table II] Progressive training (PT) is never ablated in isolation; its effect is visible only by comparing rows #5 and #6, and the direction is mixed (LSD and LSD-HF worsen from 2.326/2.568 to 2.475/2.729). Please discuss whether PT is intended to trade spectral metrics for time-domain metrics.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the PPI prior is a training-set statistic, not a test-label input, and the reported gains are empirical rather than forced by construction.

full rationale

The paper's derivation chain is a supervised mapping from 67 macroscopic plasma-state channels to 116 divertor heat-flux channels. The physical prior-aware initialization (PPI) in Eq. (4) computes a per-channel training-set mean of the target heat-flux labels and injects it into the target placeholder features via Eq. (7). This is a legitimate training-set statistic: it is fixed after training, computed only from training shots, and never uses test labels. The model is not forced to output the prior by construction; Table II shows that the prior improves MSE, but the full model additionally benefits from input perturbation and spectral loss, and the final output is still a learned function of the observed signals. The spectral-aware loss uses ground-truth spectra only as training supervision, which is standard practice. The comparison baselines are external published methods, and the self-citations (Refs. [8], [13], [34]) appear in related-work or methodological context and do not carry the central benchmark claim. The absence of a constant-prior baseline is a legitimate benchmark-completeness concern, but it is a correctness/experimental-design issue, not circularity: the paper never claims the prior alone constitutes reconstruction, and no equation identifies the prediction with the prior or with the training labels by construction.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The paper's central claim rests on the accuracy of the IR-inversion labels, the sufficiency of the 67 input signals, the independence of sampled windows, and the representativeness of 10 test shots. In addition, the method introduces a label-derived prior (116 per-channel means) and several hand-set hyperparameters; no new physical entities are introduced.

free parameters (5)
  • per-channel heat-flux priors mu_o = 116 values, one per target radial channel (Eq. 4)
    Computed as the training-set average of target labels; injected into the model input. This is a label-derived statistic and may trivially improve test performance if test shots resemble training shots. The paper does not compare against a prior-only baseline.
  • Gaussian perturbation ratio r = 0.2 (20%)
    Randomly selected fraction of observable channels perturbed; chosen by hand via a small study (Table IV).
  • Gaussian perturbation noise sigma = 0.03
    Standard deviation of noise added to selected channels; chosen by hand.
  • STFT scale weights lambda_r = not reported (R=3 scales)
    Weights for the multi-scale STFT loss (Eq. 11); values not disclosed, making exact reproduction impossible.
  • Input window length T = 500 time steps (0.5 s)
    Sliding window length; fixed by design.
assumptions (4)
  • domain assumption Post-shot IR-based heat-flux inversion provides accurate ground-truth labels
    The supervised labels are the output of conventional heat-conduction inversion; any systematic error in that inversion is inherited by the model. Section V-A.
  • domain assumption Macroscopic plasma-state signals are sufficient to determine the divertor heat-flux profile
    The entire task premise. Not proven; the paper only shows correlation on one device. Section III.
  • domain assumption Training windows are independent samples
    Windows are sampled with stride 1 from continuous shots, so neighboring windows are nearly identical; treating them as independent inflates effective sample size. Section V-A.
  • domain assumption The 10 test shots are representative of the operating space
    Only 77 shots from EAST; no evidence of coverage of the full divertor heat-flux envelope. Section V-A.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SafeDivertor: Faithful Divertor Heat Flux Reconstruction from Macroscopic Plasma State Signals via Time-Frequency Prior Exploitation." pith.science (2026). https://pith.science/paper/W73DZMHO

@misc{pith2026260805669,
  author       = {Pith},
  title        = {Pith review of: SafeDivertor: Faithful Divertor Heat Flux Reconstruction from Macroscopic Plasma State Signals via Time-Frequency Prior Exploitation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W73DZMHO}},
  note         = {Machine review of arXiv:2608.05669}
}
read the original abstract

Divertor heat-flux analysis is essential for understanding plasma-wall interactions and protecting plasma-facing components in magnetic-confinement fusion devices, while conventional infrared-based inversion is usually performed after discharge and requires heat-conduction modeling with device-specific material properties, divertor geometry, and boundary conditions. Rather than accelerating this conventional infrared-based inversion paradigm, we introduce a new online-oriented signal-based reconstruction paradigm that directly reconstructs time-resolved radial heat-flux profiles from multi-source macroscopic plasma-state signals available during discharge. To enable systematic study of this task, we construct \textbf{DivMPS2HF}, a multi-source discharge dataset that provides the data foundation and benchmark for signal-based divertor heat-flux reconstruction. We further propose \textbf{SafeDivertor}, a task-driven framework designed to address the key challenges of signal-based heat-flux reconstruction. It employs physical prior-aware initialization to provide radial-distribution guidance for target channels, input perturbation to reduce over-reliance on specific heterogeneous signals, spectral-aware reconstruction optimization to exploit time-frequency priors and preserve transient dynamics, and progressive training to stabilize the optimization of these complementary objectives. Experiments on DivMPS2HF demonstrate that SafeDivertor achieves the best overall performance among the evaluated time-series baselines across all five metrics, establishing a new performance benchmark for signal-based divertor heat-flux reconstruction. The source code will be released on https://github.com/Event-AHU/OpenFusion

Figures

Figures reproduced from arXiv: 2608.05669 by the authors.

Figure 1
Figure 1. Illustration of the conventional IR-based heat-flux inversion workflow and the proposed signal-based neural reconstruction workflow. (a) Tokamak [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of the proposed SafeDivertor framework for divertor heat-flux reconstruction from macroscopic plasma-state signals. SafeDivertor integrates [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Qualitative visualization of four randomly selected samples from different discharge shots, showing the ground truth, SafeDivertor, and T1 from top [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of four samples, one from each of four test shots. Each column corresponds to one sample, while the rows show the ground [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 31 canonical work pages

  1. [1]

    The development of a three-dimensional finite element method code for the heat flux analysis of tungsten monoblock divertor on east,

    Z. Yang, P. He, H. Yan, Z. Bin, S. Shuangbao, F. Wang, M. Chen, G. Jia, and X. Gong, “The development of a three-dimensional finite element method code for the heat flux analysis of tungsten monoblock divertor on east,”Fusion Engineering and Design, vol. 152, p. 111448, 2020

    work page 2020

  2. [2]

    Physics basis for the first iter tungsten divertor,

    R. A. Pitts, X. Bonnin, F. Escourbiac, H. Frerichs, J. Gunn, T. Hirai, A. Kukushkin, E. Kaveeva, M. Miller, D. Moultonet al., “Physics basis for the first iter tungsten divertor,”Nuclear Materials and Energy, vol. 20, p. 100696, 2019

    work page 2019

  3. [3]

    Physics informed neural networks towards the real-time calculation of heat fluxes at w7-x,

    E. Aymerich, F. Pisano, B. Cannas, G. Sias, A. Fanni, Y . Gao, D. B¨ockenhoff, M. Jakubowskiet al., “Physics informed neural networks towards the real-time calculation of heat fluxes at w7-x,”Nuclear Materials and Energy, vol. 34, p. 101401, 2023

    work page 2023

  4. [4]

    Energy flux to the asdex-upgrade diverter plates determined by thermography and calorimetry,

    A. Herrmann, W. Junker, K. Gunther, S. Bosch, M. Kaufmann, J. Neuhauser, G. Pautasso, T. Richter, and R. Schneider, “Energy flux to the asdex-upgrade diverter plates determined by thermography and calorimetry,”Plasma Physics and Controlled Fusion, vol. 37, no. 1, pp. 17–29, 1995

    work page 1995

  5. [5]

    Thermographic studies of outer target heat fluxes on kstar,

    H. Lee, R. Pitts, C. Kang, S. Oh, J. Bak, S. Hong, H. Wi, Y . Kim, H. Kim, D. Seoet al., “Thermographic studies of outer target heat fluxes on kstar,”Nuclear Materials and Energy, vol. 12, pp. 541–547, 2017

    work page 2017

  6. [6]

    Heat flux on east divertor plate in h-mode with lhcd/lhcd+ nbi,

    B. Shi, Z.-D. Yang, B. Zhang, C. Yang, K.-F. Gan, M.-W. Chen, J.-H. Yang, H. Zhang, J.-L. Qi, X.-Z. Gonget al., “Heat flux on east divertor plate in h-mode with lhcd/lhcd+ nbi,”Chinese Physics Letters, vol. 34, no. 9, p. 095201, 2017

    work page 2017

  7. [7]

    Convolutional neural networks for heat flux model validation on nstx-u,

    T. Looby, M. Reinke, D. Donovan, T. Gray, M. Messineo, and A. Kho- dak, “Convolutional neural networks for heat flux model validation on nstx-u,”IEEE Transactions on Plasma Science, vol. 48, no. 1, pp. 3–13, 2019

    work page 2019

  8. [8]

    Revisiting heat flux analysis of tungsten monoblock divertor on east us- ing physics-informed neural network,

    X. Wang, Z. Yan, H. Si, Z. Yang, Q. Yang, D. Sun, W. Lyu, and J. Tang, “Revisiting heat flux analysis of tungsten monoblock divertor on east us- ing physics-informed neural network,”arXiv preprint arXiv:2508.03776, 2025

    arXiv 2025

Show all 35 references
  1. [9]

    Predicting disruptive instabilities in controlled fusion plasmas through deep learning,

    J. Kates-Harbeck, A. Svyatkovskiy, and W. Tang, “Predicting disruptive instabilities in controlled fusion plasmas through deep learning,”Nature, vol. 568, no. 7753, pp. 526–531, 2019

    2019

  2. [10]

    Disruption prediction using a full convolutional neural network on east,

    B. Guo, B. Shen, D. Chen, C. Rea, R. Granetz, Y . Huang, L. Zeng, H. Zhang, J. Qian, Y . Sunet al., “Disruption prediction using a full convolutional neural network on east,”Plasma Physics and Controlled Fusion, vol. 63, no. 2, p. 025008, 2021

    2021

  3. [11]

    Disruption prediction for future tokamaks using parameter-based transfer learning,

    W. Zheng, F. Xue, Z. Chen, D. Chen, B. Guo, C. Shen, X. Ai, N. Wang, M. Zhang, Y . Dinget al., “Disruption prediction for future tokamaks using parameter-based transfer learning,”Communications Physics, vol. 6, no. 1, p. 181, 2023

    2023

  4. [12]

    Classification of tokamak plasma confinement states with convolutional recurrent neural networks,

    F. Matos, V . Menkovski, F. Felici, A. Pau, F. Jenko, T. Team, and E. M. Team, “Classification of tokamak plasma confinement states with convolutional recurrent neural networks,”Nuclear Fusion, vol. 60, no. 3, p. 036022, 2020

    2020

  5. [13]

    Hgts-former: Hierarchical hypergraph transformer for multivariate time series analysis,

    H. Si, X. Wang, F. Zhang, X. Zhou, D. Sun, W. Lyu, Q. Yang, and J. Tang, “Hgts-former: Hierarchical hypergraph transformer for multivariate time series analysis,”arXiv preprint arXiv:2508.02411, 2025

    2025

  6. [14]

    Tokamak edge localized mode onset prediction with deep neural network and pedestal turbulence,

    S. Joung, D. R. Smith, G. McKee, Z. Yan, K. Gill, J. Zimmerman, B. Geiger, R. Coffee, F. O’Shea, A. Jalalvandet al., “Tokamak edge localized mode onset prediction with deep neural network and pedestal turbulence,”Nuclear Fusion, vol. 64, no. 6, p. 066038, 2024

    2024

  7. [15]

    Time series extrinsic regression for reconstructing missing electron temperature in tokamak,

    M. Wang, C. Wan, J. Lu, Z. Yu, B. Xiao, Y . Li, X. He, Z. Luo, Q. Yuan, Y . Huet al., “Time series extrinsic regression for reconstructing missing electron temperature in tokamak,”Nuclear Fusion, vol. 65, no. 7, p. 076008, 2025

    2025

  8. [16]

    Deep learning-based time series forecasting: X. song et al

    X. Song, L. Deng, H. Wang, Y . Zhang, Y . He, and W. Cao, “Deep learning-based time series forecasting: X. song et al.”Artificial Intelli- gence Review, vol. 58, no. 1, p. 23, 2024

    2024

  9. [17]

    A comprehensive survey of deep learning for time series forecasting: architectural diversity and open challenges,

    J. Kim, H. Kim, H. Kim, D. Lee, and S. Yoon, “A comprehensive survey of deep learning for time series forecasting: architectural diversity and open challenges,”Artificial Intelligence Review, vol. 58, no. 7, p. 216, 2025

    2025

  10. [18]

    Deep learning for time series forecasting: a survey,

    J. F. Torres, D. Hadjout, A. Sebaa, F. Mart´ınez- ´Alvarez, and A. Troncoso, “Deep learning for time series forecasting: a survey,”Big data, vol. 9, no. 1, pp. 3–21, 2021

    2021

  11. [19]

    itrans- former: Inverted transformers are effective for time series forecasting,

    Y . Liu, T. Hu, H. Zhang, H. Wu, S. Wang, L. Ma, and M. Long, “itrans- former: Inverted transformers are effective for time series forecasting,” inInternational conference on learning representations, vol. 2024, 2024, pp. 11 116–11 140

    2024

  12. [20]

    Timer-xl: Long- context transformers for unified time series forecasting,

    Y . Liu, G. Qin, X. Huang, J. Wang, and M. Long, “Timer-xl: Long- context transformers for unified time series forecasting,” inInternational Conference on Learning Representations, vol. 2025, 2025, pp. 83 982– 84 006. IEEE TRANSACTIONS ON *** 2026 11

    2025

  13. [21]

    SimpleTM: A simple baseline for multivariate time series forecasting,

    H. Chen, V . Luong, L. Mukherjee, and V . Singh, “SimpleTM: A simple baseline for multivariate time series forecasting,” inThe Thirteenth International Conference on Learning Representations, 2025

    2025

  14. [22]

    Timefilter: Patch-specific spatial-temporal graph filtration for time series forecasting,

    Y . Hu, G. Zhang, P. Liu, D. Lan, N. Li, D. Cheng, T. Dai, S.-T. Xia, and S. Pan, “Timefilter: Patch-specific spatial-temporal graph filtration for time series forecasting,” inForty-second International Conference on Machine Learning, 2025

    2025

  15. [23]

    Frequency-domain mlps are more effective learners in time series forecasting,

    K. Yi, Q. Zhang, W. Fan, S. Wang, P. Wang, H. He, N. An, D. Lian, L. Cao, and Z. Niu, “Frequency-domain mlps are more effective learners in time series forecasting,”Advances in neural information processing systems, vol. 36, pp. 76 656–76 679, 2023

    2023

  16. [24]

    Film: Frequency improved legendre memory model for long-term time series forecasting,

    T. Zhou, Z. Ma, Q. Wen, L. Sun, T. Yao, W. Yin, R. Jinet al., “Film: Frequency improved legendre memory model for long-term time series forecasting,”Advances in neural information processing systems, vol. 35, pp. 12 677–12 690, 2022

    2022

  17. [25]

    Timesnet: Temporal 2d-variation modeling for general time series analysis,

    H. Wu, T. Hu, Y . Liu, H. Zhou, J. Wang, and M. Long, “Timesnet: Temporal 2d-variation modeling for general time series analysis,” inThe Eleventh International Conference on Learning Representations, 2023

    2023

  18. [26]

    Timeperceiver: An encoder-decoder framework for generalized time-series forecasting,

    J. Lee and H. Lee, “Timeperceiver: An encoder-decoder framework for generalized time-series forecasting,”Advances in Neural Information Processing Systems, vol. 38, pp. 136 136–136 165, 2026

    2026

  19. [27]

    Sonnet: Spectral operator neural network for multivariable time series forecasting,

    Y . Shu and V . Lampos, “Sonnet: Spectral operator neural network for multivariable time series forecasting,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 40, no. 30, 2026, pp. 25 419– 25 427

    2026

  20. [28]

    Timexer: Empowering transformers for time series forecasting with exogenous variables,

    Y . Wang, H. Wu, J. Dong, G. Qin, H. Zhang, Y . Liu, Y . Qiu, J. Wang, and M. Long, “Timexer: Empowering transformers for time series forecasting with exogenous variables,”Advances in Neural Information Processing Systems, vol. 37, pp. 469–498, 2024

    2024

  21. [29]

    GCGNet: Graph-consistent generative network for time series forecasting with exogenous variables,

    Z. Li, X. Qiu, Y . Zhu, X. Wu, J. Hu, C. Guo, and B. Yang, “GCGNet: Graph-consistent generative network for time series forecasting with exogenous variables,” inThe Fourteenth International Conference on Learning Representations, 2026

    2026

  22. [30]

    T1: One-to-one channel-head binding for multivariate time-series imputation,

    D. Park, H. Ryu, S. Bae, K. Park, and H.-S. Kim, “T1: One-to-one channel-head binding for multivariate time-series imputation,” inThe Fourteenth International Conference on Learning Representations, 2026

    2026

  23. [31]

    Saits: Self-attention-based imputation for time series,

    W. Du, D. C ˆot´e, and Y . Liu, “Saits: Self-attention-based imputation for time series,”Expert Systems with Applications, vol. 219, p. 119619, 2023

    2023

  24. [32]

    Brits: Bidirectional recurrent imputation for time series,

    W. Cao, D. Wang, J. Li, H. Zhou, L. Li, and Y . Li, “Brits: Bidirectional recurrent imputation for time series,”Advances in neural information processing systems, vol. 31, 2018

    2018

  25. [33]

    Image quality assessment: from error visibility to structural similarity,

    Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,”IEEE transactions on image processing, vol. 13, no. 4, pp. 600–612, 2004

    2004

  26. [34]

    Audio super-resolution with latent bridge models,

    C. Li, Z. Chen, L. Wang, and J. Zhu, “Audio super-resolution with latent bridge models,”Advances in Neural Information Processing Systems, vol. 38, pp. 48 593–48 636, 2026

    2026

  27. [35]

    Crosslinear: Plug-and- play cross-correlation embedding for time series forecasting with exoge- nous variables,

    P. Zhou, Y . Liu, J. Liang, Q. Song, and X. Li, “Crosslinear: Plug-and- play cross-correlation embedding for time series forecasting with exoge- nous variables,” inProceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V . 2, 2025, pp. 4120–4131

    2025

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.