Machine Learning · arXiv:2611.16789 v1 · submitted 12 Nov 2026 · 29 references
ScrewSense: Mapping the Invisible - Anomaly Detection and Error Classification in Screwdriving Processes
The paper says
“The growing volume of process data generated by screwdriving systems, combined with advancements in computing power, data storage, and algorithms, is facilitating the application of artificial intelligence in this field.”
more of the abstract
Accurately detecting and classifying screwdriving errors is crucial for ensuring product quality, avoiding costly rework in manufacturing, and supporting sustainability by facilitating the maintenance, reuse, and recycling of screw-connected products. In this context, machine learning approaches offer a promising alternative to traditional monitoring techniques, such as window-based methods or envelope curves. This study examines the use of deep learning techniques for anomaly detection and error classification in screwdriving processes. The data corpus consists of multivariate time series capturing process parameters, such as torque and rotation angle. For anomaly detection, the embedding space of a vector quantized variational autoencoder was analyzed. In terms of error classification, four different model architectures were evaluated. The results demonstrate the effectiveness of anomaly detection and confirm that deep learning models can distinguish between multiple error categories beyond traditional binary classification. This extends the capabilities of conventional monitoring approaches, achieving accuracies of up to 100%. Additionally, real-world data material was examined in order to further demonstrate model robustness. Overall the results indicate strong generalization capability and highlight the practical potential of the developed models for use in industrial screwdriving processes, eliminating the need for expert knowledge in error identification and significantly reducing manual effort and analysis time.
from the abstract
Pith says
Machine review · deepseek-v4-flash · 13 Nov 2026 · what was read was not recorded · not checked by a person
It can be wrong. Challenge this review or reply to it.
Holds up, with conditions. Solid application paper with a real gap in the anomaly detection evaluation; the error classification part is fine. This line is wrong
4 major objections · 7 minor
How often objections from reviews like this one are right: not yet measured
Not checked: figures, code, what the 29 references say.
The paper says, at §5.1, Experiment-1: Anomaly Detection
The review did not quote this passage.
Pith says: Objection 1 of 4 · major
The anomaly detection evaluation is entirely qualitative: the authors inspect t-SNE plots and manually label anomalies (Figures 2, 3, 5, 6). No quantitative metric (e.g., AUC-ROC, precision-recall, F1-score at a threshold) is reported. The abstract and conclusion claim 'effectiveness of anomaly detection', but this claim is unsupported without a quantitative measure.
more of objection 1
The authors should either provide a quantitative evaluation (e.g., using a held-out test set with known anomaly labels, or reporting separation metrics) or explicitly limit the claim to qualitative visualization. This is a load-bearing weakness because the paper's title and central claim include anomaly detection.
The paper says, at §5.2, Table 2
The review did not quote this passage.
Pith says: Objection 2 of 4 · major
Several datasets have very small sample sizes (e.g., dism c has 50 OK, 101 INCORRECT DRIVE, 50 TIGHT SCREW, 100 DAMAGED DRIVE; total ~301 samples). With 5-fold cross-validation, each fold has ~60 test samples. The reported BAC of 1.0000 on assm a and dism b raises concerns about potential overfitting or data leakage.
more of objection 2
The authors should discuss the possibility of overfitting given the small sample sizes and report confidence intervals or standard deviations across folds. Additionally, the 70/15/15 split with early stopping may lead to optimistic estimates if the validation set is used for early stopping and then the test set is evaluated; the 5-fold cross-validation is on the full dataset after the split? Clarify the exact evaluation protocol.
The paper says, at §5.4, Experiment-4: Generalization to Real-World Data
The review did not quote this passage.
Pith says: Objection 3 of 4 · major
The real-world data (36 samples) are unlabeled, so the evaluation is again qualitative: visual inspection of curves and embedding plots. The authors claim 'accurate classification of all samples based on visual inspection' (b134), but without ground truth labels, this is subjective.
more of objection 3
The model trained on dism b (which has classes REFERENCE, REFERENCE LARGE, INCORRECT DRIVE, TIGHT SCREW, DAMAGED HEAD LARGE) is applied to lawn mower data. It is unclear whether the real-world data actually contain only REFERENCE and DAMAGED HEAD LARGE classes, or whether other classes might be present but not detected. The authors should acknowledge that this is a demonstration of feasibility, not a quantitative generalization test.
The paper says, at §4.1, Anomaly Detection
The review did not quote this passage.
Pith says: Objection 4 of 4 · major
The VQVAE is pre-trained on ImageNet and fine-tuned on GAF images of normal-class samples. However, the paper does not specify the fine-tuning procedure (e.g., which layers are fine-tuned, learning rate, number of epochs). More importantly, the anomaly detection relies on visual inspection of t-SNE embeddings of the encoder outputs. t-SNE is a stochastic dimensionality reduction technique that can create spurious clusters; the authors do not discuss this limitation or provide any stability analysis (e.g., multiple runs with different random seeds).
more of objection 4
The claim that anomalies are 'detected' based on cluster separation in t-SNE plots is not rigorous.
7 minor objections
The paper says, at §3.1, Table 1
Pith says: minor objection 1 of 7
The class distribution table is difficult to parse because the column headers are not aligned with the data. Consider reformatting the table for clarity.
The paper says, at §5.2, Table 2
Pith says: minor objection 2 of 7
The table reports BAC only; the text says F1-scores were similar. It would be helpful to include F1-scores in the table or in a supplementary table for completeness.
show 5 more
The paper says, at §5.3, Table 3
Pith says: minor objection 3 of 7
The improvement on reasm a from 0.7812 to 0.9234 is attributed to correcting mislabeled samples. This is a data cleaning step, not a model improvement. The authors should clarify that the adapted dataset is a corrected version, and the comparison is not a fair evaluation of model performance but rather shows the impact of label noise.
The paper says, at §4.2, Error Classification
Pith says: minor objection 4 of 7
The description of ConvTimeNet (CTN) and ConvTran (CTR) is very brief. Since these are less common architectures, a bit more detail (or a reference to the original paper) would help readers understand the key differences.
The paper says, at §3.2, Data Preprocessing
Pith says: minor objection 5 of 7
The resampling to fixed length using linear interpolation may distort the signal if the original time series have very different lengths. The authors should discuss the choice of fixed length and its impact on the signals.
The paper says, at §5.1, Figure 2 and 3
Pith says: minor objection 6 of 7
The t-SNE plots are small and the color legend is not provided in the manuscript (the caption mentions 'colors indicating class labels' but the figure is grayscale in the arXiv version). Ensure that the figures are legible in print and that a legend is included.
The paper says, at §6, Conclusion
Pith says: minor objection 7 of 7
The conclusion states 'deep learning techniques can detect anomalies and classify errors' but the anomaly detection part is not quantitatively validated. The authors should temper this claim or explicitly note the qualitative nature of the anomaly detection evaluation.
What would settle it
Compute a quantitative anomaly detection metric on the dism a dataset: use the VQVAE reconstruction error (or a distance in embedding space) as an anomaly score, then report AUC-ROC or average precision for distinguishing normal (OK/REFERENCE) from known error classes. If AUC is near 0.5, the qualitative clusters are misleading; if AUC > 0.9, the claim is supported. This test is feasible with the existing model and data.
from the skeptic's reading
How the review went
4 readings by deepseek-v4-flash · 13 Nov 2026
- Reader holds up, with conditions
- Skeptic partly agrees
- Editor wrote the verdict above
- Referee major revision: 4 major objections, 7 minor
Not run on this paper: the simulated authors' defence.
Pith did not check this review's objections against the paper.
The review in full
The review in brief, machine-written
Hey,
This is a straightforward application paper: take standard deep learning architectures, apply them to screwdriving process data, show they work. The dataset is public, the experimental setup is reasonable (5-fold CV, early stopping, weighted loss), and the error classification results are credible. The ViT-GAF model hitting 0.945 mean BAC across five datasets is a solid result for a manufacturing application. The real-world generalization test (lawn mower data) is a nice touch, even if it's only qualitative.
What's actually new: the application domain and the specific dataset. The methods are off-the-shelf. That's fine for a conference paper, but the novelty ceiling is low.
The soft spot is the anomaly detection part. The stress-test note is right: Experiment 1 is purely qualitative. The authors show t-SNE plots, point at clusters, and call anomalies. No AUC, no precision-recall, no threshold-based metric. The abstract and conclusion claim "effectiveness of anomaly detection," but that claim is unsupported by any quantitative evaluation. The authors don't flag this as a limitation either. For a paper that markets itself as doing both anomaly detection and error classification, this is a load-bearing gap. The error classification experiments are fine, but the anomaly detection half is incomplete.
Other concerns are minor. The 100% accuracy on some datasets (assm a) is suspicious but not impossible given the small sample sizes and simple task. The authors acknowledge the small sample issue indirectly. The lack of code is a reproducibility concern, but the dataset is public.
Who this is for: researchers in manufacturing monitoring who want a baseline application. It's a reasonable conference paper for CIRP CMS. The error classification results are worth reporting; the anomaly detection needs a quantitative follow-up.
Recommendation: I'd send this to peer review. The error classification work is solid enough to warrant referee time, and the anomaly detection gap is fixable with a revision. The paper is not groundbreaking, but it's honest applied work with a public dataset.
The strongest objection
The paper's central claim includes that deep learning can detect anomalies in screwdriving processes. For anomaly detection, the authors use a VQVAE embedding space visualized via t-SNE and manually inspect clusters (Figures 2, 3, 5, 6). No quantitative anomaly detection metric (e.g., AUC-ROC, precision-recall, F1-score at a threshold) is reported. The evaluation is entirely qualitative: the authors look at scatter plots and label points as anomalous based on visual separation. This is a load-bearing weakness because the claim of 'effectiveness' for anomaly detection cannot be assessed without a quantitative measure. The reader's weakest_assumption correctly identifies that the VQVAE embedding space may not reliably separate anomalies, but the deeper issue is that the paper never actually tests that separation quantitatively. The authors acknowledge limitations (reliance on labeled data, transferability), but do not flag the absence of a quantitative anomaly detection evaluation. The error classification experiments (Experiments 2-4) do have quantitative metrics, but the anomaly detection component (Experiment 1) remains unvalidated. Since the abstract and conclusion claim effectiveness for anomaly detection, this gap undermines a core part of the central claim.
The reader's weakest_assumption focuses on the reliability of the VQVAE embedding space for separation, which is related. However, the more fundamental issue is that the paper never quantitatively evaluates anomaly detection at all. The reader's verdict is CONDITIONAL, which is appropriate given this gap. I agree with CONDITIONAL but for a more specific reason: the anomaly detection claim is unsubstantiated by any metric. The error classification part is quantitatively evaluated and seems plausible, so the overall verdict remains CONDITIONAL pending a quantitative anomaly detection evaluation. The paper's own sentences (b7, b144) claim effectiveness for anomaly detection, but no quantitative evidence is provided. The authors do not state this as a limitation; they present the qualitative analysis as sufficient. This is not an attack on the authors' integrity, but a concrete missing piece in the evidence chain. The concrete test I propose is straightforward and could be done with the existing model and data, and would settle whether the anomaly detection claim holds.
The referee's report
If the results hold, the paper demonstrates a practical application of deep learning to an industrial monitoring problem, potentially reducing manual inspection effort. The comparison of multiple architectures on a new public dataset is a useful contribution. However, the significance is limited by the small dataset sizes (around 50 samples per class) and the qualitative nature of the anomaly detection evaluation. The claim of 'effectiveness' for anomaly detection is not quantitatively supported, which weakens the overall contribution. The error classification results are more solid but still rely on small datasets where 100% accuracy may indicate overfitting or lack of challenge. The paper acknowledges limitations (reliance on labeled data, transferability, lack of benchmarks) but does not address the absence of quantitative anomaly detection metrics.
Pith's reading of the paper (a machine's summary, not the authors')
Screwdriving is a common manufacturing process where the quality of each screw joint must be checked. Traditionally, this is done by setting fixed limits on torque and angle, which can miss subtle errors. This paper explores using deep learning to automatically detect anomalies and classify different types of errors from the torque and angle time series data. The authors use a vector quantized variational autoencoder (VQVAE) to learn a compressed representation of normal screwdriving curves. By projecting these representations into a 2D space using t-SNE, they can visually identify anomalies that deviate from the normal pattern. They also train four different classification models to distinguish between error types: a multilayer perceptron (MLP), a convolutional neural network (CNN), a transformer, and a vision transformer (ViT) that operates on images created from the time series data. The best performing model is the ViT with Gramian Angular Field (GAF) transformation, which converts the time series into images. It achieves an average balanced accuracy of 94.5% across five datasets. On some datasets, it reaches 100% accuracy. The model also generalizes to real-world data from lawn mower housings, correctly classifying 33 out of 36 samples as normal and identifying three damaged screws. The paper demonstrates that deep learning can effectively monitor screwdriving processes, potentially reducing the need for expert manual inspection. However, the study has limitations: the datasets are relatively small (around 50 samples per class), the models may not transfer well to different screwdriving systems, and the anomaly detection is qualitative rather than quantitative.
Core claim. Deep learning techniques can detect anomalies and classify errors in the evaluation of screwdriving processes, achieving accuracies of up to 100% and generalizing to real-world data.
The premise it rests on. The assumption that the VQVAE embedding space, trained only on normal class samples, can reliably separate anomalies from normal samples in screwdriving processes, and that the ViT-GAF model trained on one dataset (dism b) can generalize to real-world data without significant domain shift.
Check this review
Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.
How often reviews like this one are right
Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.
How often objections from reviews like this one are right: not yet measured
Reviews by people
This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.
The paper
Abstract
The growing volume of process data generated by screwdriving systems, combined with advancements in computing power, data storage, and algorithms, is facilitating the application of artificial intelligence in this field. Accurately detecting and classifying screwdriving errors is crucial for ensuring product quality, avoiding costly rework in manufacturing, and supporting sustainability by facilitating the maintenance, reuse, and recycling of screw-connected products. In this context, machine learning approaches offer a promising alternative to traditional monitoring techniques, such as window-based methods or envelope curves. This study examines the use of deep learning techniques for anomaly detection and error classification in screwdriving processes. The data corpus consists of multivariate time series capturing process parameters, such as torque and rotation angle. For anomaly detection, the embedding space of a vector quantized variational autoencoder was analyzed. In terms of error classification, four different model architectures were evaluated. The results demonstrate the effectiveness of anomaly detection and confirm that deep learning models can distinguish between multiple error categories beyond traditional binary classification. This extends the capabilities of conventional monitoring approaches, achieving accuracies of up to 100%. Additionally, real-world data material was examined in order to further demonstrate model robustness. Overall the results indicate strong generalization capability and highlight the practical potential of the developed models for use in industrial screwdriving processes, eliminating the need for expert knowledge in error identification and significantly reducing manual effort and analysis time.
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.
abstract compiled · 2026-11-13
The authors' main claim, as the classifier picked it out 0.98
“Overall the results indicate strong generalization capability and highlight the practical potential of the developed models for use in industrial screwdriving processes, eliminating the need for expert knowledge in error identification and significantly reducing manual effort and analysis time.”
How the main claim is stated: hedged possibility 0.80
What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.09 · human participants 0.04 · controlled comparison 0.19 · randomized assignment 0.02 · states quantitative result 0.98 · compares with prior results 0.48 · mentions code or data availability 0.03
The abstract, sentence by sentence
This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.
| Sentence | Role | The authors' own work | States the main conclusion | Strength |
|---|---|---|---|---|
| The growing volume of process data generated by screwdriving systems, combined with advancements in computing power, data storage, and algorithms, is facilitating the application of artificial intelligence in this field. | background 1.00 | 0.06 | 0.02 | asserted 0.97 |
| Accurately detecting and classifying screwdriving errors is crucial for ensuring product quality, avoiding costly rework in manufacturing, and supporting sustainability by facilitating the maintenance, reuse, and recycling of screw-connected products. | background 1.00 | 0.05 | 0.03 | asserted 1.00 |
| In this context, machine learning approaches offer a promising alternative to traditional monitoring techniques, such as window-based methods or envelope curves. | background 0.95 | 0.05 | 0.03 | hedged possibility 0.84 |
| This study examines the use of deep learning techniques for anomaly detection and error classification in screwdriving processes. | contribution 0.99 | 0.80 | 0.03 | asserted 0.80 |
| The data corpus consists of multivariate time series capturing process parameters, such as torque and rotation angle. | method 1.00 | 0.81 | 0.02 | asserted 1.00 |
| For anomaly detection, the embedding space of a vector quantized variational autoencoder was analyzed. | method 1.00 | 0.92 | 0.02 | asserted 0.96 |
| In terms of error classification, four different model architectures were evaluated. | method 0.98 | 0.92 | 0.02 | asserted 0.97 |
| The results demonstrate the effectiveness of anomaly detection and confirm that deep learning models can distinguish between multiple error categories beyond traditional binary classification. | result 0.94 | 0.95 | 0.52 | asserted 1.00 |
| This extends the capabilities of conventional monitoring approaches, achieving accuracies of up to 100%. | result 0.78 | 0.89 | 0.19 | asserted 0.99 |
| Additionally, real-world data material was examined in order to further demonstrate model robustness. | method 0.81 | 0.92 | 0.03 | asserted 0.98 |
| Overall the results indicate strong generalization capability and highlight the practical potential of the developed models for use in industrial screwdriving processes, eliminating the need for expert knowledge in error identification and significantly reducing manual effort and analysis time. | interpretation 0.96 | 0.89 | 0.84 | hedged possibility 0.80 |
What kind of paper the classifier reports this is: method or tool 0.50 · evidence from observation or experiment 0.67
What was examined, and what was not
Sources read
- arXiv listing abstract v1
Not examined
- full text
- figures and tables
- references and citing sentences
- relations to other papers
Levels run
- P abstract
Not run
- S sections
- T body sentences
- C citations
- R cross-paper relations
How the labels were made
Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-13: 1 request, 8,141 input tokens.
Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON
Figures (7)
Showing 6 of the paper's 7 figures. Figures are the authors', reproduced from the paper.
References (29)
- [1] N. Rabbe, Messtechnik in der Schraubtechnik, in: R. Schmitt, E. Diet- rich (Eds.), Handbuch Messtechnik in der industriellen Produktion: Valide Messergebnisse planen, erhalten, auswerten und verteilen, 1st Edition, Carl Hanser Verlag GmbH & Co. KG, M¨unchen, 2023, pp. 359–374 on Pith
- [2] DEPRAG SCHULZ GMBH u. CO. KG, Technical Information: Screw- driving Technology and Quality Assurance, Tech. rep., DEPRAG SCHULZ GMBH u. CO. KG (2020). URLhttps://www.deprag.com/fileadmin/bilder_content/ emedia/broschueren_pics/Allgemein_PDF/D0080/D0080en.pdf on Pith
- [3] WEBER Schraubautomaten GmbH, Schraubtechnik Ratgeber (2025). URLhttps://www.weber-online.com/ schraubtechnik-ratgeber/ on Pith
- [4] W.-D. Schneider, H ¨ullkurvenverfahren zur Prozeߨuberwachung, in: Flexi- bel automatisiertes Taumelnieten, Springer Berlin Heidelberg, Berlin, Hei- delberg, 1997, pp. 58–65.doi:10.1007/978-3-662-07247-9_7 on Pithdoi:10.1007/978-3-662-07247-9_7
- [5] Nord-Lock Group, Benefits of Using Bolted Joints in the Industrial Sector (2021). URLhttps://www.nord-lock.com/learnings/bolting-tips/ 2021/industrial-bolted-joints/ on Pith
- [6] K. Althoefer, B. Lara, Y . Zweiri, L. D. Seneviratne, Automated failure clas- sification for assembly with self-tapping threaded fastenings using artificial neural networks, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Scien... on Pithdoi:10.1243/09544062jmes546
- [7] M. Meiners, A. Mayr, J. Franke, Process curve analysis with machine learn- ing on the example of screw fastening and press-in processes, Procedia CIRP 97 (2021) 166–171.doi:10.1016/j.procir.2020.05.220 on Pithdoi:10.1016/j.procir.2020.05.220
- [8] Anwar Al Assadi and David Holtz and Frank N ¨agele and Christof Nitsche and Werner Kraus and Marco F. Huber, Machine learning based screw drive state detection for unfastening screw connections, Journal of Manu- facturing Systems 65 (2022) 19–32.doi:https://doi.org/10.1016/ j.... on Pith
- [9] pytorch.org, torch.nn.functional.interpolate. URLhttps://pytorch.org/docs/stable/generated/torch.nn. functional.interpolate.html on Pith
- [10] A. Asesh, Normalization and Bias in Time Series Data, in: C. Biele, J. Kacprzyk, W. Kope ´c, J. W. Owsi ´nski, A. Romanowski, M. Sikorski (Eds.), Digital Interaction and Machine Intelligence, Springer International Publishing, Cham, 2022, pp. 88–97 on Pith
- [11] Y . Wu, K. He, Group normalization, in: Proceedings of the European con- ference on computer vision (ECCV), 2018, pp. 3–19 on Pith
- [12] Q. Wen, L. Sun, F. Yang, X. Song, J. Gao, X. Wang, H. Xu, Time Series Data Augmentation for Deep Learning: A Survey, in: Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI- 2021, International Joint Conferences on Artificial Intellige... on Pithdoi:10.24963/ijcai.2021/631
- [13] numpy.org, numpy.histogram2d. URLhttps://numpy.org/doc/stable/reference/generated/ numpy.histogram2d.html on Pith
- [14] Z. Wang, T. Oates, Encoding Time Series as Images for Visual Inspection and Classification Using Tiled Convolutional Neural Networks, 2015 on Pith
- [15] pytorch.org, torch.stft. URLhttps://pytorch.org/docs/stable/generated/torch. stft.html on Pith
- [16] C. Bergler, Deep Learning Applied to Animal Linguistics, Phd thesis, Friedrich-Alexander-Universit ¨at Erlangen-N ¨urnberg (FAU), available athttps://nbn-resolving.org/urn:nbn:de:bvb: 29-opus4-239268(9 2023) on Pith
- [17] A. van den Oord, O. Vinyals, K. Kavukcuoglu, Neural Discrete Represen- tation Learning (2018).arXiv:1711.00937. URLhttps://arxiv.org/abs/1711.00937 on PithPith's reviewarXiv
- [18] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, A. C. Berg, L. Fei-Fei, Ima- geNet Large Scale Visual Recognition Challenge, International Journal of Computer Vision (IJCV) 115 (3) (2015) 211–252.doi:10.1007/ s1126... on Pith
- [19] L. Van der Maaten, G. Hinton, Visualizing data using t-SNE., Journal of machine learning research 9 (11) (2008) on Pith
- [20] V . Ekambaram, A. Jati, N. Nguyen, P. Sinthong, J. Kalagnanam, TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecast- ing, in: Proceedings of the 29th ACM SIGKDD Conference on Knowl- edge Discovery and Data Mining, KDD ’23, ACM, 2023, p. 459–469. doi:10.11... on Pithdoi:10.1145/3580305.3599533
show 9 more
- [21] Q. Wen, T. Zhou, C. Zhang, W. Chen, Z. Ma, J. Yan, L. Sun, Transformers in Time Series: A Survey (2023).arXiv:2202.07125. URLhttps://arxiv.org/abs/2202.07125 on PithPith's reviewarXiv
- [22] M. Cheng, J. Yang, T. Pan, Q. Liu, Z. Li, ConvTimeNet: A Deep Hierar- chical Fully Convolutional Model for Multivariate Time Series Analysis (2024).arXiv:2403.01493. URLhttps://arxiv.org/abs/2403.01493 on PithPith's reviewarXiv
- [23] S. Islam, H. Elmekki, A. Elsebai, J. Bentahar, N. Drawel, G. Rjoub, W. Pedrycz, A Comprehensive Survey on Applications of Transformers for Deep Learning Tasks (2023).arXiv:2306.07303. URLhttps://arxiv.org/abs/2306.07303 on PithPith's reviewarXiv
- [24] A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Un- terthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, N. Houlsby, An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (2021).arXiv:2010.11929. URLhttps://arxiv.or... on PithPith's reviewarXiv
- [25] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. u. Kaiser, I. Polosukhin, Attention is All you Need, in: I. Guyon, U. V . Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, R. Garnett (Eds.), Advances in Neural Information Processing Systems... on Pith
- [26] S. Talukder, Y . Yue, G. Gkioxari, Totem: Tokenized time series embeddings for general time series analysis, arXiv preprint arXiv:2402.16412 (2024) on PithPith's reviewarXiv
- [27] L. Sun, J. Liang, C. Zhang, D. Wu, Y . Zhang, Meta-Transfer Metric Learn- ing for Time Series Classification in 6G-Supported Intelligent Transporta- tion Systems, IEEE Transactions on Intelligent Transportation Systems 25 (3) (2024) 2757–2767.doi:10.1109/TITS.2023.3250962 on Pithdoi:10.1109/tits.2023.3250962
- [28] J. Schuster, A. W ¨olfel, F. Brunner, C. Bergler, PredTrAD – Prediction- based Transformer for Anomaly Detection in Multivariate Time Series Data, in: Interspeech 2025, 2025, pp. 3873–3877.doi:10.21437/ Interspeech.2025-501 on Pith
- [29] N. West, J. Deuse, PyScrew: A Comprehensive Dataset Collection from Industrial Screw Driving Experiments (2025).arXiv:2505.11925. URLhttps://arxiv.org/abs/2505.11925 on PithPith's reviewarXiv
Challenges and discussion
Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.
Come back by email
A short list of the week's newly reviewed papers. Nothing is sent until you confirm from your inbox, and every email has a link that stops it. No account.
No comments yet.
Sign in to comment