Pith. sign in

REVIEW 5 major objections 6 minor 32 references

ViSIR: Vision Transformer Single Image Reconstruction Method for Earth System Models

T0 review · 5 major / 6 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read A hybrid vision transformer with sinusoidal output layers reconstructs upscaled Earth-system-model images with less error than four established super-resolution methods.

desk verdict The PSNR gains are not supported because the paper never specifies a held-out split, so the reported comparisons look like training-set fit; the architecture idea is plausible but the evaluation needs a full redo. read the letter →

arxiv 2502.06741 v3 pith:IZT2L4T6 submitted 2025-02-10 cs.CV

Ehsan Zeraatkar , Salah Faroughi , Jelena Tešić This is my paper · ORCID
classification cs.CV MSC 68T0768U10
keywords singleimagesuper-resolutionEarthsystemmodelsimplicitneuralrepresentationSIRENvisiontransformerspectralbiasclimatedatadownscalingE3SM
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

ViSIR is a proposed architecture for single-image super-resolution (SR) of Earth system model (ESM) outputs. It combines the global-context modeling of a Vision Transformer (ViT) with the high-frequency detail recovery of a Sinusoidal Representation Network (SIREN), replacing the transformer's final fully connected layer with a SIREN layer. On a dataset of 540 images derived from E3SM fields (surface temperature, shortwave heat flux, longwave heat flux), the authors report that ViSIR achieves higher PSNR and SSIM and lower MSE than SRCNN, ViT, SIREN, and SRGAN. If the reported gains hold on data unseen during training, ViSIR would offer a practical way to store coarse ESM output and reconstruct fine-resolution fields for regional climate analysis.

What carries the argument

The central object is the ViSIR architecture, which replaces the final fully connected layer of a Vision Transformer with a SIREN layer whose output is HR = sin(ω0 · (W F + b)), where F is the aggregated token feature vector and ω0 is a trainable-scale frequency hyperparameter. The mechanism is a hyperparameter search over ω0 (10 to 60 Hz) and SIREN hidden layers (1 to 6), selecting ω0 = 20 with two hidden layers based on mean PSNR. This integration is designed to combine ViT's patch-based global-context modeling with SIREN's high-frequency representation strength to combat spectral bias in super-resolution.

What would settle it

Retrain ViSIR and all baselines on a documented training split of the 540-image E3SM dataset and evaluate on a held-out month of images, reporting PSNR on the held-out set; if the margin over SRCNN shrinks below 2 dB or reverses, the claimed superiority of ViSIR as a generalizer fails.

Watch

Extended reading notes

Core claim

The paper claims that a Vision Transformer whose final reconstruction layer is replaced by a SIREN layer, with a tuned frequency parameter ω0 = 20 and two hidden layers, outperforms four baselines on the ESM super-resolution task. Specifically, the abstract reports average PSNR improvements of 2.16 dB over SRCNN, 6.29 dB over ViT, 8.34 dB over SIREN, and 7.93 dB over SRGAN for three measured ESM variables. The authors attribute the gain to mitigating spectral bias: ViT captures long-range dependencies while SIREN's periodic activations preserve high-frequency details that standard networks tend to smooth away.

Load-bearing premise

The load-bearing premise is that the reported PSNR, SSIM, and MSE numbers were computed on images not used during training or hyperparameter selection, but the paper never describes a train/test split.

Editorial extensions

If this is right

  • If the reported gains generalize to held-out ESM data, coarse-resolution climate model outputs could be stored at lower cost and reconstructed at 4x resolution with lower error than existing SR methods.
  • The hybrid architecture provides a template for other SR tasks where both global context and sharp local detail matter, such as satellite imagery and weather downscaling.
  • The explicit frequency tuning over ω0 suggests that choosing the right periodic activation scale is a transferable design decision for implicit neural representation based SR models.
  • The reported SSIM improvements imply that the reconstructed fields preserve structural and contrast information better than the baselines, which matters for downstream climate analyses that depend on spatial gradients.

Reading between the lines

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

  • The frequency-search methodology implies that the optimal ω0 may vary across climate variables or grid resolutions, so a fixed ω0 = 20 might be suboptimal for other ESM fields or finer upsampling factors; this is an extension the paper does not test.
  • Because the paper only evaluates 4x upsampling of images already derived from interpolated model grids, a natural next test is whether ViSIR also beats baselines on raw non-interpolated cubed-sphere grid data or on higher upsampling factors such as 8x.
  • If the hybrid architecture is extended to video or multi-image SR, the SIREN's continuous functional form could allow querying the reconstructed field at arbitrary spatial locations, which the paper leaves implicit.
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

5 major / 6 minor

Summary. The paper proposes ViSIR, a hybrid architecture that appends Sinusoidal Representation Network (SIREN) layers to a Vision Transformer (ViT) for 4x single-image super-resolution of E3SM climate model data. The authors construct a 540-image RGB dataset from three ESM variables, run a hyperparameter search over SIREN frequency and layer count, and report max/mean/min MSE, PSNR, and SSIM against SRCNN, ViT, SIREN, and SRGAN. The abstract claims average PSNR improvements of 2.16 dB over SRCNN, 6.29 dB over ViT, 8.34 dB over SIREN, and 7.93 dB over SRGAN. The paper also lists an ablation study as a main contribution and concludes that ViSIR produces negligible reconstruction error.

Significance. If the empirical claims were established, the ViT+SIREN combination would be a reasonable contribution to a practical SR task for Earth system model downscaling, and the application domain is relevant. The paper offers a clear problem statement and a simple architecture that is easy to describe. However, the evaluation as reported does not establish held-out performance: no train/test/validation split is stated, the baseline comparison protocol is unfair to at least one baseline, no ablation is performed despite being claimed, and the reported numbers are internally inconsistent. The paper provides no code, no data split, no training loss, no optimization details, and no uncertainty estimates, so the central performance claim is not currently reproducible or auditable.

major comments (5)
  1. [Sections V and VI] The paper never specifies a train/test/validation split. Section V describes 540 images (10 months x 18 tiles x 3 variables) but does not state how they are partitioned for training and testing. Table 1 reports max/mean/min over the dataset for each method, and Figure 2 shows a hyperparameter search over 180 Surface Temperature images; Section VI then states that the best parameters are used for the rest of the methods. As written, the most parsimonious reading is that all models were trained and evaluated on the same 540 images, so the reported PSNR/SSIM advantages in the Abstract may reflect fitting or memorization rather than generalization. Without a held-out test set, the central claim that ViSIR outperforms the baselines is not established.
  2. [Section VI] The baseline comparison is not fair. The SIREN baseline is forced to use the hyperparameters tuned for ViSIR (omega_0=20 and two hidden layers) to perform a fair comparison, which can handicap SIREN if its own optimal configuration is different. No hyperparameter search or tuning is reported for SRCNN, ViT, or SRGAN, and no training details (loss function, optimizer, learning rate, epochs, batch size) are given for any method. The reported margins over SIREN and SRGAN are therefore not interpretable as clean architecture effects.
  3. [Section I.A and Section VI] An ablation study is listed as a main contribution ("Ablation Study of Model Components"), but no ablation experiment appears in Section VI. The paper only compares the full ViSIR model against independent baselines; it never isolates the contribution of the ViT backbone, the SIREN module, or the integration design. This claimed contribution is unsupported by the presented experiments.
  4. [Abstract, Table 1, and Section VI.A] The numerical results are internally inconsistent. Averaging the per-variable mean PSNR differences from Table 1 gives a ViSIR-ViT gain of about 6.42 dB, not the 6.29 dB stated in the Abstract. Section VI.A claims gains of 10.6 dB over SIREN and 7.8 dB over ViT, neither of which matches the mean or max values in Table 1. Section VII claims improvements of up to 9.93 dB and up to 4.29 dB that are not reproducible from the table. Section VI.B's statement that "the ViSIR best PSNR is 36.7% better than ViT" is not a meaningful operation on dB values. These inconsistencies make it impossible to audit the reported superiority.
  5. [Table 1 and Figure 4] Table 1 and Figure 4 report max, mean, and min values over the dataset, but no standard deviations, confidence intervals, or repeated-run statistics are provided. With a single run per method and no significance testing, the reported differences may be within noise. The paper needs at least error bars or a significance test before claiming consistent improvement across the three variables.
minor comments (6)
  1. [Section II, last paragraph] The text says the work builds on advances "by integrating CNNs, transformers, and innovative loss functions," but the proposed method does not integrate CNNs and no loss function is described anywhere in the paper; the narrative should be aligned with the actual method.
  2. [Section VIII] The conclusion begins with "INIn," which appears to be a typo and should be corrected.
  3. [Equations (3) and (4)] Equation (3) defines f(x) = sin(omega_0 W x + b) but does not define the input x in the SIREN context, and Equation (4) reuses W and b without specifying their dimensions relative to the transformer output T; the notation should be clarified.
  4. [Algorithm 1] Algorithm 1, line 10 applies a SIREN network inside each transformer block, while Section III and Equation (4) place SIREN after the transformer output; the pipeline should be made unambiguous.
  5. [Table 1] The header uses "Source Temperature" while the text and Figure 3 use "Surface Temperature"; the terminology should be consistent.
  6. [General] No code, data link, or full training configuration is provided, which prevents reproduction of the reported experiments.

Circularity Check

1 steps flagged · score 6.0 of 10

The reported ViSIR superiority is a fitted result: hyperparameters are selected on the same Surface Temperature images used for the headline comparison, and no held-out split is described.

  1. fitted input called prediction [Section VI Experimental Results, Figure 2 caption and Table 1; Section V dataset description]
    "We applied the hyper-parameter searches with different sinusoidal activation function frequencies ranging from 10 to 60 Hz while varying the number of hidden layers in the SIREN from 1 to 6 to get the best hyper-parameters. ... The best PSNR value for EMS images is two hidden layers with a frequency of 20. These are the hyper-parameters used for the rest of the methods in frequency and layers to perform a fair comparison. [Figure 2 caption:] ... PSNR values ... applied to 180 images of the Surface Temperature variable."

    Figure 2 chooses omega0=20 and two SIREN hidden layers by maximizing mean PSNR on the 180 Surface Temperature images (10 months x 18 tiles). Table 1 then reports max/mean/min PSNR for ViSIR on the Source Temperature measurement, i.e., the same 180 images, and the abstract converts those numbers into the claim that ViSIR outperforms SRCNN, ViT, SIREN, and SRGAN. Section V states only that the dataset contains 540 images; no train/validation/test split is ever described. Thus the 'best parameters' were chosen using exactly the data on which the comparative metrics are reported. The ViSIR entries in Table 1 are therefore a product of fitting the configuration to the evaluation set, not an independent prediction of held-out reconstruction quality.

full rationale

The paper contains no mathematical derivation chain whose conclusion is identical to its premises; the circularity is in the empirical loop. The sole load-bearing circular step is the hyperparameter tuning/evaluation overlap: the frequency and hidden-layer count are optimized on the same 180 Surface Temperature images that appear in Table 1's Source Temperature row, and no held-out split is documented anywhere in Sections V or VI. Consequently the abstract's headline improvements are fitted values rather than demonstrated predictions. This warrants a 6 rather than a 0-2 because the central claim (ViSIR outperforms four baselines) is the quantity affected by the fit, not a peripheral detail. The self-citation to [14] provides the dataset and prior ESM SR comparisons but is not itself load-bearing for the score, since the E3SM data are external. The internal arithmetic inconsistencies and missing ablation are correctness/reporting issues, not additional circularity.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on an under-specified empirical pipeline: synthetic downsampling, a small dataset, undisclosed training details, and an unstated test split. The free parameters are hyperparameters chosen on the same data. No code, further data, or independent falsifiable predictions are provided.

free parameters (3)
  • SIREN frequency omega_0 = 20
    Chosen via hyperparameter search on the ESM dataset (Section VI, Figure 2) and applied to all methods.
  • Number of SIREN hidden layers = 2
    Chosen via the same search and used in ViSIR and the baseline SIREN.
  • ViT architecture hyperparameters (patch size, layers, heads, embedding dimension) = not reported
    Algorithm 1 lists these as inputs but their values and the training configuration (learning rate, epochs, optimizer, loss) are never stated.
assumptions (4)
  • domain assumption Bicubic interpolation from a 0.25 degree grid to a 1 degree grid produces valid low-resolution versions of ESM fields for training SR models.
    Section V constructs LR images via bicubic interpolation; this assumes the blur-downsample model matches the real ESM resolution loss.
  • ad hoc to paper Comparing all methods under the same SIREN hyperparameters (omega_0=20, two hidden layers) is a fair comparison.
    Section VI says the searched parameters are used 'for the rest of the methods in frequency and layers to perform a fair comparison.' This may handicap the SIREN baseline.
  • domain assumption The reported metrics were computed on data not used for training.
    No train/test split is described; this unstated assumption underlies the validity of Table 1.
  • domain assumption Encoding three ESM variables as RGB channels is a meaningful image representation for SR.
    Section V derives R, G, B from normalized temperature and heat fluxes; this assumes cross-variable color encoding helps or at least does not hurt SR.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ViSIR: Vision Transformer Single Image Reconstruction Method for Earth System Models." pith.science (2026). https://pith.science/paper/IZT2L4T6

@misc{pith2026250206741,
  author       = {Pith},
  title        = {Pith review of: ViSIR: Vision Transformer Single Image Reconstruction Method for Earth System Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IZT2L4T6}},
  note         = {Machine review of arXiv:2502.06741}
}
read the original abstract

Purpose: Earth system models (ESMs) integrate the interactions of the atmosphere, ocean, land, ice, and biosphere to estimate the state of regional and global climate under a wide variety of conditions. The ESMs are highly complex; thus, deep neural network architectures are used to model the complexity and store the down-sampled data. This paper proposes the Vision Transformer Sinusoidal Representation Networks (ViSIR) to improve the ESM data's single image SR (SR) reconstruction task. Methods: ViSIR combines the SR capability of Vision Transformers (ViT) with the high-frequency detail preservation of the Sinusoidal Representation Network (SIREN) to address the spectral bias observed in SR tasks. Results: The ViSIR outperforms SRCNN by 2.16 db, ViT by 6.29 dB, SIREN by 8.34 dB, and SR-Generative Adversarial (SRGANs) by 7.93 dB PSNR on average for three different measurements. Conclusion: The proposed ViSIR is evaluated and compared with state-of-the-art methods. The results show that the proposed algorithm is outperforming other methods in terms of Mean Square Error(MSE), Peak-Signal-to-Noise-Ratio(PSNR), and Structural Similarity Index Measure(SSIM).

Figures

Figures reproduced from arXiv: 2502.06741 by the authors.

Figure 1
Figure 1. FIGURE 1 [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIGURE 2 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. FIGURE 3 [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: FIGURE 4 [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: FIGURE 5 [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: FIGURE 6 [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 26 canonical work pages

  1. [1]

    Heavens, D

    N. Heavens, D. Ward, and M. M. Natalie, ‘‘Studying and projecting climate change with earth system models,’’ Nature Education Knowledge , vol. 4, no. 5, 2013

    work page 2013

  2. [2]

    Heinze, V

    C. Heinze, V . Eyring, P . Friedlingstein, C. Jones, Y . Balkanski, W. Collins, T. Fichefet, S. Gao, A. Hall, D. Ivanova et al. , ‘‘Esd reviews: Climate feedbacks in the earth system and prospects for their evaluation,’’ Earth Syst. Dynam. , vol. 10, pp. 379–452, 2019. [Online]. Available: https://doi.org/10.5194/esd-10-379-2019

    work page doi:10.5194/esd-10-379-2019 2019

  3. [3]

    Eyring, S

    V . Eyring, S. Bony, G. A. Meehl, C. A. Senior, B. Stevens, R. J. Stouffer, and K. E. Taylor, ‘‘Overview of the coupled model intercomparison project phase 6 (cmip6) experimental design and organization,’’ Geoscientific Model Development , vol. 9, no. 5, pp. 1937–1958, 2016. [Online]. Available: https://doi.org/10.5194/gmd-9-1937-2016

    doi:10.5194/gmd-9-1937-2016 1937

  4. [5]

    DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolution

    [Online]. Available: https://arxiv.org/abs/1703.03126

    work page Pith review arXiv

  5. [6]

    Y ang, X

    W. Y ang, X. Zhang, Y . Tian, W. Wang, J. Xue, and Q. Liao, ‘‘Deep learning for single image super-resolution: A brief review,’’ IEEE Transactions on Multimedia, vol. 21, no. 12, pp. 3106–3121, 2019

    work page 2019

  6. [7]

    C. Dong, C. C. Loy, K. He, and X. Tang, ‘‘Image super-resolution using deep convolutional networks,’’IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, no. 2, pp. 295–307, 2016

    work page 2016

  7. [8]

    Ledig, L

    C. Ledig, L. Theis, F. Huszár, J. Caballero, A. Cunningham, A. Acosta, A. Aitken, A. Tejani, J. Totz, Z. Wang, and W. Shi, ‘‘Photo-realistic single image super-resolution using a generative adversarial network,’’ in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , 2017, pp. 4681–4690. [Online]. Available: https://arxi...

    arXiv 2017

  8. [9]

    Y . Tai, J. Y ang, X. Liu, and C. Xu, ‘‘Memnet: A persistent memory network for image restoration,’’ in Proceedings of the IEEE International Conference on Computer Vision (ICCV) , 2017, pp. 4539–4547. [Online]. Available: https://arxiv.org/abs/1708.02209

    work page Pith review arXiv 2017

Show all 32 references
  1. [10]

    Touvron, M

    H. Touvron, M. Cord, M. Douze, F. Massa, A. Sablayrolles, and H. Jégou, ‘‘Training data-efficient image transformers & distillation through attention,’’ in Proceedings of the 38th International Conference on Machine Learning (ICML), 2021. [Online]. Available: https://arxiv.org...

    2021 arXiv

  2. [11]

    Dosovitskiy, L

    A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Un- terthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, and N. Houlsby, ‘‘An image is worth 16x16 words: Transformers for image recognition at scale,’’ ArXiv, vol. abs/2010.11929, 2020. [...

    2010 arXiv

  3. [12]

    J. Bai, L. Y uan, S.-T. Xia, S. Y an, Z. Li, and W. Liu, ‘‘Improving vision transformers by revisiting high-frequency components,’’ in Proceedings of the European Conference on Computer Vision (ECCV) , 2022

    2022

  4. [13]

    Sitzmann, J

    V . Sitzmann, J. N. P . Martel, A. W. Bergman, D. B. Lindell, and G. Wet- zstein, ‘‘Implicit neural representations with periodic activation functions,’’ in Proceedings of the 34th International Conference on Neural Information Processing Systems, ser. NIPS ’20. Red Hook, NY ,...

    2020

  5. [14]

    Zhang, Z

    X. Zhang, Z. Zhang, S. Wu, and Z. Zhang, ‘‘Residual networks behave like ensembles of relatively shallow networks,’’ IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 41, no. 6, pp. 1311–1325, 2019

    2019

  6. [15]

    N. M. Pawar, R. Soltanmohammadi, S. K. Mahjour, and S. A. Faroughi, ‘‘Esm data downscaling: a comparison of super-resolution deep learning models,’’ Earth Science Informatics , pp. 1–18, 2024

    2024

  7. [16]

    J. Kim, J. K. Lee, and K. M. Lee, ‘‘Enhanced deep residual networks for single image super-resolution,’’ Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 136–144, 2016

    2016

  8. [17]

    1646–1654

    ——, ‘‘Accurate image super-resolution using very deep convolutional networks,’’ in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 1646–1654

    2016

  9. [18]

    Zhang, Y

    Y . Zhang, Y . Tian, Y . Kong, B. Zhong, and Y . Fu, ‘‘Residual dense network for image super-resolution,’’ in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 2472–2481

    2018

  10. [19]

    Grattarola and P

    D. Grattarola and P . V andergheynst, ‘‘Generalised implicit neural represen- tations,’’ in Proceedings of the 36th International Conference on Neural Information Processing Systems , ser. NIPS ’22. Red Hook, NY , USA: Curran Associates Inc., 2022

    2022

  11. [20]

    Y . Chen, R. Wu, Y . Liu, and C. Zhu, ‘‘Hoin: High-order implicit neural representations,’’ 2024. [Online]. Available: https://arxiv.org/abs/2404. 14674

    2024

  12. [21]

    Shidqi, C

    N. Shidqi, C. Jeong, S. Park, E. Zeller, A. B. Nellikkattil, and K. Singh, ‘‘Generating high-resolution regional precipitation using conditional diffusion model,’’ 2023. [Online]. Available: https://arxiv.org/abs/2312. 07112

    2023

  13. [22]

    S. Ding, X. Zhi, Y . Lyu, Y . Ji, and W. Guo, ‘‘Deep learning for daily 2-m temperature downscaling,’’ Earth and Space Science , vol. 11, 02 2024

    2024

  14. [23]

    Irani and V

    H. Irani and V . Metsis, ‘‘Positional encoding in transformer-based time series models: A survey,’’ arXiv preprint arXiv:2502.12370, 2025

    2025 arXiv

  15. [24]

    Zhong, F

    X. Zhong, F. Du, L. Chen, Z. Wang, and H. Li, ‘‘Investigating transformer- based models for spatial downscaling and correcting biases of near-surface temperature and wind-speed forecasts,’’ Quarterly Journal of the Royal Meteorological Society, vol. 150, pp. 275–289, 2023

    2023

  16. [25]

    Karwowska and D

    K. Karwowska and D. Wierzbicki, ‘‘Modified esrgan with uformer for video satellite imagery super-resolution,’’Remote Sensing, vol. 16, p. 1926, 2024

    1926

  17. [26]

    C. Ma, J. Mi, W. Gao, and S. Tao, ‘‘Desrgan: Detail-enhanced generative adversarial networks for small sample single image super-resolution,’’ Neurocomputing, vol. 617, p. 129121, 2025

    2025

  18. [27]

    W. Guo, P . Lu, X. Peng, and Z. Zhao, ‘‘Learnable adaptive bilateral filter for improved generalization in single image super-resolution,’’ Pattern Recognition, vol. XX, pp. XX–XX, 2025

    2025

  19. [28]

    Wu and Y

    C. Wu and Y . Jing, ‘‘Unsupervised super resolution using dual contrastive learning,’’ Neurocomputing, vol. 630, p. 129649, 2025

    2025

  20. [29]

    V . S. F. Garnot, L. Landrieu, S. Giordano, and N. Chehata, ‘‘Satellite image time series classification with pixel-set encoders and temporal self- attention,’’ in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 12 325–12 334

    2021

  21. [30]

    Y ang, H

    F. Y ang, H. Y ang, J. Fu, H. Lu, and B. Guo, ‘‘Learning texture transformer network for image super-resolution,’’ in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 2020, pp. 5791–5800

    2020

  22. [31]

    Carion, F

    N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, and S. Zagoruyko, ‘‘End-to-end object detection with transformers,’’ in Proceedings of the European Conference on Computer Vision (ECCV) , 2020, pp. 213–229. [Online]. Available: https://arxiv.org/abs/2005.12872

    2020 arXiv

  23. [32]

    E3SM Project, ‘‘Energy Exascale Earth System Model (E3SM),’’ [Com- puter Software], Mar. 2024

    2024

  24. [33]

    L. S. Passarella, S. Mahajan, A. Pal, and M. R. Norman, ‘‘Reconstructing high-resolution esm data through a novel fast super-resolution convolu- tional neural network (fsrcnn),’’Earth and Space Science, vol. 49, 02 2022. 7

    2022

Pith tools

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