Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

DenoMAE: A Multimodal Autoencoder for Denoising Modulation Signals

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read DenoMAE reaches 83.5% modulation-classification accuracy from 10,000 unlabeled and 1,000 labeled samples by treating noise as an explicit input modality.

desk verdict Plausible application of MultiMAE to modulation signals, but the headline data-efficiency claims are unsupported by an uncontrolled comparison table. read the letter →

arxiv 2501.11538 v1 pith:A3M36PRA submitted 2025-01-20 cs.LG

Atik Faysal , Taha Boushine , Mohammad Rostami , Reihaneh Gh. Roshan , Huaxia Wang , Nikhil Muralidhar , Avimanyu Sahoo , Yu-Dong Yao This is my paper · ORCID
classification cs.LG
keywords multi-modalityvisiontransformermodulationclassificationconstellationdiagramsdenoisingmaskedautoencoderself-supervisedpretraininglowSNR
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 asks whether a self-supervised autoencoder can clean noisy radio signals well enough that modulation classification requires almost no labeled data. The proposed answer, DenoMAE, pre-trains a masked autoencoder on five views of the same signal—noisy and noiseless waveforms, noisy and noiseless constellation diagrams, and an explicit noise view—so the model must reconstruct the clean views from the noisy ones. The paper reports that this pretraining lets a ten-way modulation classifier reach 83.5% accuracy using only 10,000 unlabeled and 1,000 labeled samples, with the largest gains at low signal-to-noise ratios and with denoising that still works below the SNR range seen in training. A sympathetic reader would take the paper's central bet to be that noise, treated as an input rather than only as corruption, is a reusable source of self-supervision for communication signals. If that bet pays off, radio-frequency classification becomes practical in settings where labeled captures are scarce and noise conditions are challenging.

What carries the argument

The load-bearing mechanism is multimodal masked reconstruction with noise as a co-equal modality. Each of the five input modalities is a three-channel 224-by-224 image, divided into 16-by-16 patches and linearly embedded; a twelve-layer transformer encoder processes only the 25% of patches left visible after random masking, a linear projection followed by layer normalization maps each modality into a shared latent space, and a four-layer decoder reconstructs all masked patches for all modalities from the concatenated shared code. The training loss is a weighted sum of mean-squared-error reconstruction losses across the five modalities. Because the decoder must regenerate the noiseless views from representations that also contain the noisy views, the shared latent space is where the model learns to attribute structure to signal and noise separately; downstream classification discards the decoder and uses only the encoder.

What would settle it

Train DenoMAE and a same-architecture classifier from scratch on the identical 10-class dataset at the same 0.5 dB SNR, giving both only 1,000 labeled samples; if the from-scratch classifier matches or exceeds 83.5% accuracy, the pretraining advantage is refuted. A second check is to retrain one of the comparison systems on DenoMAE's dataset and sample counts and see whether 10,000 pretraining and 1,000 fine-tuning examples still win.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that a masked autoencoder can be turned into a data-efficient modulation classifier by making noise a first-class input modality. DenoMAE is pre-trained on 10,000 unlabeled samples, each rendered as five paired views: a noisy constellation diagram (a scatter plot of a signal's amplitude and phase states), a noisy signal image, the corresponding noiseless constellation diagram, the noiseless signal image, and an explicit noise image. With 75% of patches randomly masked, a shared encoder must reconstruct all masked patches of all five views from the unmasked ones, which forces it to separate signal content from noise. After fine-tuning the encoder with only 1,000 labeled samples, the model reports 83.5% test accuracy across ten modulation classes at 0.5 dB SNR, 77.5% at -10 dB, a 22.1% gain over the same architecture without pretraining at that SNR, and meaningful denoising on SNRs down to -20 dB, below the pretraining range.

Load-bearing premise

The headline numbers compare DenoMAE, tested on its own generated 10-class dataset at a signal-to-noise ratio of 0.5 dB, with earlier systems evaluated on different class counts, SNRs, and sample sizes; if those cross-paper numbers cannot be fairly compared, the claimed accuracy and data-efficiency advantage has no common baseline.

Editorial extensions

If this is right

  • A modulation classifier for a previously unseen band could be built from roughly 10,000 unlabeled captures and 1,000 labeled examples, rather than the hundreds of thousands of samples used by earlier systems.
  • Pretraining matters most in the regime where supervised classifiers typically fail: at low SNR, where the reported gain over no pretraining is 22.1 percentage points at -10 dB.
  • Because the model denoises inputs at SNRs below its training range, a model pretrained on a convenient SNR window could be deployed in noisier environments without retraining.
  • The ablation result—accuracy rising from 81.3% with a single modality to 83.5% with all five—indicates that each added view, including the explicit noise view, contributes a small but consistent gain.

Reading between the lines

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

  • An implication the authors do not spell out is that the same noise-as-modality mechanism should transfer to other denoising tasks with paired clean and noisy observations, such as speech enhancement or radar clutter suppression, because nothing in the architecture is modulation-specific.
  • A fair test of the data-efficiency claim would be to hold one dataset fixed and vary only the number of pretraining and fine-tuning samples; the absolute gap might shrink, but the direction of the claim would then be established.
  • A testable extension suggested by the architecture is to vary the masking ratio per modality: if the noise view is truly load-bearing, downstream accuracy should be more sensitive to masking the noise view heavily than to masking any other single view heavily.
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 / 4 minor

Summary. The paper proposes DenoMAE, a multimodal masked-autoencoder pretraining approach for automatic modulation classification. The model operates on five input modalities (noisy constellation image, noisy signal, clean constellation image, clean signal, and noise), masks 75% of patches, and reconstructs the masked content; the pretrained encoder is then fine-tuned with a two-layer MLP head for 10-class classification. The authors report 83.5% accuracy at 0.5 dB SNR and 77.5% at -10 dB using 10,000 pretraining and 1,000 fine-tuning samples, and claim a 10x reduction in unlabeled pretraining data and a 3x reduction in labeled fine-tuning data relative to prior methods, as well as a 22.1% accuracy gain over a non-pretrained model at low SNR.

Significance. If the empirical claims were substantiated, the paper would address a relevant problem: data-efficient automatic modulation classification at low SNR, with the interesting design choice of including noise as an explicit modality. The per-modality ablation in Table I provides useful evidence that the additional modalities contribute to downstream accuracy. However, the significance of the paper rests almost entirely on the validity of Table II and on the claimed 22.1% pretraining gain, and neither is supported by the evidence as presented. The manuscript does not provide a controlled comparison, a public benchmark evaluation, error bars, or a definition of the core noise modality, so the central contribution cannot currently be assessed.

major comments (4)
  1. [Section V.E, Table II] The state-of-the-art claim is not supported by Table II: DenoMAE is evaluated on a private 10-class dataset at 0.5 dB SNR, while AlexNet uses 2 classes at 8 dB with 800,000 training samples, NMformer uses 10 classes on a different dataset at 0.5-4.5 dB, CNN-AMC uses 4 classes at -6 dB, and DL-GRF uses 4 classes at 0 dB. Accuracy in AMC depends strongly on class count, SNR, signal length, and dataset difficulty, so the 83.5% figure cannot be directly compared with these numbers. The derived claims of 80x fewer samples than AlexNet and 10.7x fewer than NMformer are therefore not meaningful without a controlled comparison on the same data, same class count, same SNR protocol, and ideally the same architecture family.
  2. [Section V.B] The claimed 22.1% improvement over the non-pretrained baseline at -10 dB SNR is unverifiable: the manuscript reports no from-scratch baseline, no architecture-matched supervised-only training result, and no table or figure for this comparison. Since the pretraining benefit is a central contribution, the authors need to include a same-architecture baseline trained from scratch on the same 1,000 labeled examples, with multiple random seeds, and preferably report this comparison at every SNR point shown in Figure 4.
  3. [Section III; Section V.A] The term unlabeled oversells the pretraining requirement. Although class labels are not used, the reconstruction targets are clean paired noiseless signals and constellation diagrams for each noisy input, so the pretraining data require paired clean versions of every sample. This is a form of signal-level supervision rather than purely unlabeled data. The data-efficiency claim should be restated to acknowledge this requirement, and the paper should explain how clean targets are obtained in a practical deployment setting.
  4. [Section III-A; Section IV-A] The noise-as-explicit-modality, which is the main novelty of the paper, is never defined. The manuscript does not specify how the noise modality is constructed, for example whether it is the residual between the noisy and clean signals, a separate noise image, or a noise-power map, and no equation or algorithm for computing it is provided. This omission prevents reproduction of the method and leaves the core contribution untestable; the authors should specify exactly how each of the five modalities is computed.
minor comments (4)
  1. [Abstract; Section I; Section V.E] The abstract says a 10% reduction in unlabeled pretraining data and a 3% reduction in labeled fine-tuning data, while the introduction and Section V.E claim one-tenth and one-third (10x and 3x) reductions; these statements are contradictory and should be corrected.
  2. [Section IV.A] The text mentions a 7x7 complex plane but the input images are 224x224; please clarify how the constellation samples are mapped to this grid and how the three exponential decay rates are chosen, since these are free parameters.
  3. [Abstract; Section V.A] The statement that our codes are public at GitHubDenoMAE does not include a URL or repository identifier; a complete link should be provided.
  4. [Table I; Figure 4] All accuracy numbers are point estimates; please report means and standard deviations over multiple runs and include error bars in the figures.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: DenoMAE's central claims are empirical, and the main concerns (unmatched baselines, 'unlabeled' pretraining) are validity and interpretation issues, not derivation-by-construction.

full rationale

The paper contains no formal derivation chain that reduces to its own inputs. DenoMAE is an empirical architecture plus a reconstruction objective: the loss is L = sum_m w_m L_m with L_m an MSE between reconstructed and target patches. The model is trained to map noisy input modalities toward noiseless reconstruction targets; this is standard supervised denoising, not a case where a predicted quantity is defined in terms of the quantity it is supposed to predict. Calling the pretraining 'unlabeled' while using noiseless modalities as reconstruction targets is a semantic overstatement, but it does not make the result circular: the reconstruction target is not statistically forced by the input alone, and the downstream classification accuracy is a separate empirical measurement. The claimed '22.1% improvement over the non-pre-trained baseline' is not shown, and Table II compares against methods on different datasets, class counts, and SNRs; those are serious evidential weaknesses for the state-of-the-art and data-efficiency claims, but they are not circularity. There is no imported uniqueness theorem, no ansatz smuggled via citation that carries the argument, and no renaming of a known result as organization. The one overlapping-author reference ([34]) is used only as an external baseline in a comparison table; it is not load-bearing for the method's derivation. Accordingly, the circularity score is 0.

Assumptions & free parameters 5 free parameters · 4 assumptions · 1 invented entities

The central claims rest on several unverified modeling choices: clean paired targets during pretraining, a comparability assumption across incomparable baselines, and unspecified constellation rendering and loss-weight values. These are not fitted constants in a derivation but hand-chosen ingredients that affect every reported accuracy.

free parameters (5)
  • Mask ratio p_mask = 0.75
    Hand-selected masking fraction used in pretraining; no sensitivity study is reported.
  • Modality loss weights w_m = Not reported
    The total loss is a weighted sum over five modalities, but the weights are never specified; they determine the pretraining objective.
  • Constellation decay rates alpha = Not reported
    The three RGB channels are generated with three distinct exponential decay rates, but the values are omitted; this choice defines the input representation.
  • Number of modalities = 5
    The paper's central design choice; no ablation across 2-4 modalities except a single additive table.
  • SNR training range = -10 dB to 10 dB
    The claimed extrapolation below -10 dB is only measured by visual inspection, and the training SNR distribution is not described.
assumptions (4)
  • domain assumption Paired noisy and noiseless versions of the same signal are available for every pretraining sample.
    Reconstruction targets in Section III-A5 are noiseless signals and diagrams, so pretraining is not label-free in the strict sense; clean targets must exist.
  • ad hoc to paper Baselines in Table II are comparable across different datasets, SNRs, classes, and training sizes.
    The SOTA and data-efficiency claims depend on these cross-table comparisons being meaningful, which is not established.
  • domain assumption A 224x224 three-channel constellation image retains enough information for modulation classification after denoising.
    The entire pipeline uses resized and interpolated constellation images; no analysis shows this representation is sufficient or optimal.
  • domain assumption Multimodal masked pretraining with a shared latent space transfers from natural images to radio signal modalities.
    The architecture is borrowed from MultiMAE without theoretical or empirical justification for radio signals beyond the final accuracy numbers.
invented entities (1)
  • Noise-as-modality
    purpose: A fifth input modality representing noise is fed through the encoder so the model can reconstruct clean signals and diagrams from noisy inputs.
    This is the paper's main conceptual contribution, but the construction of the noise image is not specified and no falsifiable prediction outside the paper's dataset is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DenoMAE: A Multimodal Autoencoder for Denoising Modulation Signals." pith.science (2026). https://pith.science/paper/A3M36PRA

@misc{pith2026250111538,
  author       = {Pith},
  title        = {Pith review of: DenoMAE: A Multimodal Autoencoder for Denoising Modulation Signals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A3M36PRA}},
  note         = {Machine review of arXiv:2501.11538}
}
read the original abstract

We propose Denoising Masked Autoencoder (Deno-MAE), a novel multimodal autoencoder framework for denoising modulation signals during pretraining. DenoMAE extends the concept of masked autoencoders by incorporating multiple input modalities, including noise as an explicit modality, to enhance cross-modal learning and improve denoising performance. The network is pre-trained using unlabeled noisy modulation signals and constellation diagrams, effectively learning to reconstruct their equivalent noiseless signals and diagrams. Deno-MAE achieves state-of-the-art accuracy in automatic modulation classification tasks with significantly fewer training samples, demonstrating a 10% reduction in unlabeled pretraining data and a 3% reduction in labeled fine-tuning data compared to existing approaches. Moreover, our model exhibits robust performance across varying signal-to-noise ratios (SNRs) and supports extrapolation on unseen lower SNRs. The results indicate that DenoMAE is an efficient, flexible, and data-efficient solution for denoising and classifying modulation signals in challenging noise-intensive environments.

Figures

Figures reproduced from arXiv: 2501.11538 by the authors.

Figure 1
Figure 1. DenoMAE Pretraining Strategy: We apply a random 75% masking (not to scale in the illustration) across all input modalities. The remaining 25% of visible patches are processed by a shared encoder, while each modality utilizes a dedicated decoder to reconstruct its masked patches. Only the encoder is reused for fine-tuning the downstream tasks. [10]. Recent advances in multimodal learning address this limitation by in… view at source ↗
Figure 2
Figure 2. Denoised outputs of DenoMAE on unlabeled constellation diagrams [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Extrapolation-ability of DenoMAE on out-of-bound much lower [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: DenoMAE fine-tuned classification accuracy for constellation dia [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Unsupervised Time-Series Signal Analysis with Autoencoders and Vision Transformers: A Review of Architectures and Applications

    cs.LG 2025-04 reject novelty 3.0 of 10

    A narrative review of autoencoder and vision transformer methods for unsupervised time-series signal analysis, with several citation mismatches in the text and tables.

Reference graph

Works this paper leans on

36 extracted references · 27 canonical work pages · cited by 1 Pith paper

  1. [1]

    A cookbook of self-supervised learning,

    R. Balestriero, M. Ibrahim, V . Sobal, A. Morcos, S. Shekhar, T. Gold- stein, F. Bordes, A. Bardes, G. Mialon, Y . Tian, A. Schwarzschild, A. G. Wilson, J. Geiping, Q. Garrido, P. Fernandez, A. Bar, H. Pirsiavash, Y . LeCun, and M. Goldblum, “A cookbook of self-supervised learning,” 2023

    work page 2023

  2. [2]

    Exploring the limits of transfer learning with a unified text-to-text transformer,

    A. Roberts, C. Raffel, K. Lee, M. Matena, N. Shazeer, P. J. Liu, S. Narang, W. Li, and Y . Zhou, “Exploring the limits of transfer learning with a unified text-to-text transformer,” Google, Tech. Rep. , 2019

    work page 2019

  3. [3]

    Emerging properties in self-supervised vision transformers,

    M. Caron, H. Touvron, I. Misra, H. J ´egou, J. Mairal, P. Bojanowski, and A. Joulin, “Emerging properties in self-supervised vision transformers,” in Proceedings of the IEEE/CVF international conference on computer vision, pp. 9650–9660, 2021

    2021

  4. [4]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark, et al., “Learning transferable visual models from natural language supervision,” in International conference on machine learning , pp. 8748–8763, PMLR, 2021

    2021

  5. [5]

    Deep learning-based snr estimation,

    S. Zheng, S. Chen, T. Chen, Z. Yang, Z. Zhao, and X. Yang, “Deep learning-based snr estimation,” IEEE Open Journal of the Communica- tions Society , 2024

    work page 2024

  6. [6]

    Adaptive modulation with cazac preamble-based signal-to-noise-ratio estimator in ofdm cooperative com- munication system,

    S. Manzoor and N. S. Othman, “Adaptive modulation with cazac preamble-based signal-to-noise-ratio estimator in ofdm cooperative com- munication system,” IEEE Access , vol. 10, pp. 126550–126560, 2022

    work page 2022

  7. [7]

    Learn- ing the unknown: Improving modulation classification performance in unseen scenarios,

    E. Perenda, S. Rajendran, G. Bovet, S. Pollin, and M. Zheleva, “Learn- ing the unknown: Improving modulation classification performance in unseen scenarios,” in IEEE INFOCOM 2021-IEEE Conference on Computer Communications , pp. 1–10, IEEE, 2021

    work page 2021

  8. [8]

    Adaptive spatial modulation mimo based on machine learning,

    P. Yang, Y . Xiao, M. Xiao, Y . L. Guan, S. Li, and W. Xiang, “Adaptive spatial modulation mimo based on machine learning,” IEEE Journal on Selected Areas in Communications , vol. 37, no. 9, pp. 2117–2131, 2019

    work page 2019

Show all 36 references
  1. [9]

    Gaf-mae: A self- supervised automatic modulation classification method based on gramian angular field and masked autoencoder,

    Y . Shi, H. Xu, Y . Zhang, Z. Qi, and D. Wang, “Gaf-mae: A self- supervised automatic modulation classification method based on gramian angular field and masked autoencoder,” IEEE Transactions on Cognitive Communications and Networking , 2023

    2023

  2. [10]

    Deep learning of radio frequency fingerprints from limited samples by masked autoencoding,

    K. Huang, H. Liu, P. Hu, et al. , “Deep learning of radio frequency fingerprints from limited samples by masked autoencoding,” IEEE Wireless Communications Letters , 2022

    2022

  3. [11]

    Multimae: Multi- modal multi-task masked autoencoders,

    R. Bachmann, D. Mizrahi, A. Atanov, and A. Zamir, “Multimae: Multi- modal multi-task masked autoencoders,” in European Conference on Computer Vision, pp. 348–367, Springer, 2022

    2022

  4. [12]

    4m-21: An any-to-any vision model for tens of tasks and modalities,

    R. Bachmann, O. F. Kar, D. Mizrahi, A. Garjani, M. Gao, D. Griffiths, J. Hu, A. Dehghan, and A. Zamir, “4m-21: An any-to-any vision model for tens of tasks and modalities,”arXiv preprint arXiv:2406.09406, 2024

    2024 arXiv

  5. [13]

    Joint variational autoen- coders for multimodal imputation and embedding,

    N. Cohen Kalafut, X. Huang, and D. Wang, “Joint variational autoen- coders for multimodal imputation and embedding,” Nature Machine Intelligence, vol. 5, no. 6, pp. 631–642, 2023

    2023

  6. [14]

    The dipencoder: Enforcing multimodality in autoencoders,

    C. Leiber, L. G. Bauer, M. Neumayr, C. Plant, and C. B ¨ohm, “The dipencoder: Enforcing multimodality in autoencoders,” in Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , pp. 846–856, 2022

    2022

  7. [15]

    Fusion methods for cnn-based automatic modulation classification,

    S. Zheng, P. Qi, S. Chen, and X. Yang, “Fusion methods for cnn-based automatic modulation classification,” IEEE Access , vol. 7, pp. 66496– 66504, 2019

    2019

  8. [16]

    Cnn-based automatic modulation classification for beyond 5g communications,

    A. P. Hermawan, R. R. Ginanjar, D.-S. Kim, and J.-M. Lee, “Cnn-based automatic modulation classification for beyond 5g communications,” IEEE Communications Letters , vol. 24, no. 5, pp. 1038–1041, 2020

    2020

  9. [17]

    Modulation classification using convolutional neural network based deep learning model,

    S. Peng, H. Jiang, H. Wang, H. Alwageed, and Y .-D. Yao, “Modulation classification using convolutional neural network based deep learning model,” in 2017 26th Wireless and Optical Communication Conference (WOCC), pp. 1–5, IEEE, 2017

    2017

  10. [18]

    Mcnet: An efficient cnn architecture for robust automatic modulation classification,

    T. Huynh-The, C.-H. Hua, Q.-V . Pham, and D.-S. Kim, “Mcnet: An efficient cnn architecture for robust automatic modulation classification,” IEEE Communications Letters , vol. 24, no. 4, pp. 811–815, 2020

    2020

  11. [19]

    Automatic modulation recognition using deep learning architectures,

    M. Zhang, Y . Zeng, Z. Han, and Y . Gong, “Automatic modulation recognition using deep learning architectures,” in 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications (SPA WC), pp. 1–5, 2018

    2018

  12. [20]

    Automatic modulation recognition using deep cvcnn-lstm architecture,

    R. Cheng, Q. Chen, and M. Huang, “Automatic modulation recognition using deep cvcnn-lstm architecture,” Alexandria Engineering Journal , vol. 104, pp. 162–170, 2024

    2024

  13. [21]

    Research on modulation recognition method in low snr based on lstm,

    B. Zhang, G. Chen, and C. Jiang, “Research on modulation recognition method in low snr based on lstm,” in Journal of Physics: Conference Series, p. 012003, IOP Publishing, 2022

    2022

  14. [22]

    Automatic modulation recognition based on cnn and gru,

    F. Liu, Z. Zhang, and R. Zhou, “Automatic modulation recognition based on cnn and gru,” Tsinghua Science and Technology , vol. 27, no. 2, pp. 422–431, 2021

    2021

  15. [23]

    A transformer-based contrastive semi-supervised learning framework for automatic modula- tion recognition,

    W. Kong, X. Jiao, Y . Xu, B. Zhang, and Q. Yang, “A transformer-based contrastive semi-supervised learning framework for automatic modula- tion recognition,” IEEE Transactions on Cognitive Communications and Networking, vol. 9, no. 4, pp. 950–962, 2023

    2023

  16. [24]

    A transformer-based ctdnn structure for automatic modulation recognition,

    W. Kong, Q. Yang, X. Jiao, Y . Niu, and G. Ji, “A transformer-based ctdnn structure for automatic modulation recognition,” in 2021 7th International Conference on Computer and Communications (ICCC) , pp. 159–163, IEEE, 2021

    2021

  17. [25]

    Signal modulation classification based on the transformer network,

    J. Cai, F. Gan, X. Cao, and W. Liu, “Signal modulation classification based on the transformer network,” IEEE Transactions on Cognitive Communications and Networking , vol. 8, no. 3, pp. 1348–1357, 2022

    2022

  18. [26]

    Bert: Pre-training of deep bidirectional transformers for language understanding,

    J. Devlin, “Bert: Pre-training of deep bidirectional transformers for language understanding,” arXiv preprint arXiv:1810.04805 , 2018

    2018 arXiv

  19. [27]

    Mae: Masked autoencoders are scalable vision learners,

    K. He, X. Chen, S. Xie, Y . Li, P. Doll ´ar, and R. Girshick, “Mae: Masked autoencoders are scalable vision learners,” arXiv preprint arXiv:2111.06377, 2021

    2021 arXiv

  20. [28]

    Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks,

    J. Lu, D. Batra, D. Parikh, and S. Lee, “Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks,” Advances in neural information processing systems , vol. 32, 2019

    2019

  21. [29]

    Uniter: Universal image-text representation learning,

    Y .-C. Chen, L. Li, L. Yu, A. El Kholy, F. Ahmed, Z. Gan, Y . Cheng, and J. Liu, “Uniter: Universal image-text representation learning,” in European conference on computer vision , pp. 104–120, Springer, 2020

    2020

  22. [30]

    Image as a foreign language: Beit pretraining for all vision and vision-language tasks,

    W. Wang, H. Bao, L. Dong, J. Bjorck, Z. Peng, Q. Liu, K. Aggarwal, O. K. Mohammed, S. Singhal, S. Som, et al. , “Image as a foreign language: Beit pretraining for all vision and vision-language tasks,” arXiv preprint arXiv:2208.10442 , 2022

    2022 arXiv

  23. [31]

    4m: Massively multimodal masked modeling,

    D. Mizrahi, R. Bachmann, O. Kar, T. Yeo, M. Gao, A. Dehghan, and A. Zamir, “4m: Massively multimodal masked modeling,” Advances in Neural Information Processing Systems , vol. 36, 2024

    2024

  24. [32]

    Learning constellation map with deep cnn for accurate modulation recognition,

    V .-S. Doan, T. Huynh-The, C.-H. Hua, Q.-V . Pham, and D.-S. Kim, “Learning constellation map with deep cnn for accurate modulation recognition,” in GLOBECOM 2020-2020 IEEE Global Communications Conference, pp. 1–6, IEEE, 2020

    2020

  25. [33]

    Deep learning for constellation-based modulation classifica- tion under multipath fading channels,

    T. Huynh-The, C.-H. Hua, V .-S. Doan, Q.-V . Pham, T.-V . Nguyen, and D.-S. Kim, “Deep learning for constellation-based modulation classifica- tion under multipath fading channels,” in 2020 International Conference on Information and Communication Technology Convergence (ICTC)...

    2020

  26. [34]

    Modulation classification based on signal constellation diagrams and deep learning,

    S. Peng, H. Jiang, H. Wang, H. Alwageed, Y . Zhou, M. M. Sebdani, and Y .-D. Yao, “Modulation classification based on signal constellation diagrams and deep learning,” IEEE transactions on neural networks and learning systems , vol. 30, no. 3, pp. 718–727, 2018

    2018

  27. [35]

    Automatic modulation classification: A deep learning enabled approach,

    F. Meng, P. Chen, L. Wu, and X. Wang, “Automatic modulation classification: A deep learning enabled approach,” IEEE Transactions on V ehicular Technology, vol. 67, no. 11, pp. 10760–10772, 2018

    2018

  28. [36]

    Automatic modulation classification using techniques from image classification,

    Y . Sun and E. A. Ball, “Automatic modulation classification using techniques from image classification,” IET Communications , vol. 16, no. 11, pp. 1303–1314, 2022

    2022

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.