Pith. sign in

REVIEW 4 major objections 5 minor 50 references

Scaling Particle Collision Data Analysis

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A task-agnostic byte-level transformer matches specialized jet-tagging models and jumps sharply past a data threshold.

desk verdict A generic byte-level transformer matching specialized jet taggers at 10M events is an interesting and plausible result, but the scaling/emergence claim is confounded with compute and needs major revision. read the letter →

arxiv 2412.00129 v2 pith:JFHUPYUX submitted 2024-11-28 cs.LG hep-exphysics.data-an

Hengkui Wu , Panpan Chi , Yongfeng Zhu , Liujiang Liu , Shuyang Hu , Yuexin Wang , Chen Zhou , Qihao Wang
show 5 more authors
Yingsi Xin Bruce Liu Dahao Liang Xinglong Jia Manqi Ruan
This is my paper · ORCID
classification cs.LGhep-exphysics.data-an
keywords LargelanguagemodelsBinarytokenizationJetoriginidentificationtaggingScalinglawsEmergenceNumericaldataParticlephysics
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 a task-agnostic transformer trained on raw bytes can match specialist deep-learning models at Jet Origin Identification, the task of telling which quark, antiquark, or gluon initiated a jet in a particle collider. The proposed model, BBT-Neutron, uses binary tokenization, feeding numbers in as native byte sequences to preserve the quantitative structure that byte-pair encoding destroys. At ten million training events, its flavor-tagging efficiencies and charge-flip rates are comparable to ParticleNet and Particle Transformer. The paper also reports a sharp, S-shaped jump in BBT-Neutron's performance once the training set crosses a critical size, a behavior it reads as evidence for emergent, generalist capabilities in numerically intensive science.

What carries the argument

The load-bearing mechanism is Binary Tokenization: every input, whether text, numbers, formulas, or images, is converted into byte sequences, with numbers optionally encoded as their native integer or float byte arrays rather than as character strings, and the bytes are grouped into fixed-length patches fed to a causal transformer decoder. Because each byte is a token and token IDs follow the byte layout, the same number always produces the same tokens and nearby numbers have nearby representations, which the paper argues preserves quantitative relationships that BPE fragments. The transformer uses rotary position embeddings and patch-level self-attention that also mixes information within a patch, letting a single left-to-right, sequence-to-sequence model act as a classifier without jet-specific structure.

What would settle it

Retrain BBT-Neutron at dataset sizes from 100 to 10 million events with early stopping or with epochs scaled by dataset size, and plot final validation accuracy against data volume; if the sharp jump smooths away or shifts, the claimed emergence is an artifact of the fixed 30-epoch budget.

Watch

Extended reading notes

Core claim

The central discovery is that domain-specific architecture is not required for state-of-the-art jet origin identification: a generic decoder-only transformer with causal attention, rotary position embeddings, and byte-level binary tokenization reaches the same 11-way classification quality as the leading specialist models once it sees enough data. The paper further claims that this generalist model shows a different scaling curve from the specialists. It performs near random guessing on small datasets, then jumps sharply above random after roughly ten thousand events for charge identification, and converges with ParticleNet and Particle Transformer around three million events. For bottom, charm, and strange quarks, BBT-Neutron needs roughly an order of magnitude more training data to match the specialists, and for bottom, strange, and up quarks its scaling curve trends above theirs at the largest sizes. The authors interpret the abrupt transition as the emergence of a generalist capability, analogous to emergent abilities in large language models, and attribute its absence in specialists to their built-in symmetry and interaction priors.

Load-bearing premise

The load-bearing premise is that one fixed training budget (30 epochs, batch size 512) is a fair comparison at every dataset size; if the small-dataset runs are merely undertrained, the sharp jump is an optimization artifact rather than an emergent transition.

Editorial extensions

If this is right

  • At ten million events, BBT-Neutron's 11-category confusion matrix, flavor-tagging efficiencies, and charge-flip rates match ParticleNet and Particle Transformer, so specialist architecture is not necessary for strong JoI performance at that scale.
  • The sharp performance jump between ten and thirty thousand events means there is a minimum data volume below which a byte-level generalist is not competitive, making dataset size a key operating parameter.
  • The same byte representation works for text, numbers, formulas, and images, so the architecture can be pretrained on a mixed corpus and then applied to classification or regression with only an output-layer change.
  • For bottom, strange, and up quarks, BBT-Neutron's scaling curves continue to improve past the point where the specialists plateau, suggesting that further data may widen the generalist's advantage.

Reading between the lines

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

  • Inference: The emergence claim rests on an unverified fixed-budget assumption; the paper does not show convergence checks or learning curves, so the sharp jump could be small-dataset undertraining rather than a phase transition.
  • Inference: Re-plotting the scaling curves with early stopping or with training epochs scaled by dataset size would separate optimization effects from true emergent behavior; if the jump persists under matched compute, the generalist claim is much stronger.
  • Inference: Because binary tokenization claims to preserve numeric relations, a cheap transfer test is to benchmark BBT-Neutron on arithmetic and number-comparison tasks; standard LLM arithmetic failures would be expected to shrink or vanish if the mechanism works as described.
  • Inference: The comparison against permutation-invariant specialists suggests a testable hypothesis: adding permutation augmentation to BBT-Neutron's left-to-right input should reduce the data threshold, showing that symmetry prior, not architecture class, drives the gap.
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

4 major / 5 minor

Summary. The paper proposes BBT-Neutron, a decoder-only transformer that uses byte-level ("binary") tokenization and patch embeddings as a task-agnostic architecture for scientific numerical data. It is evaluated on Jet Origin Identification (JoI) using simulated e+e−→νν̄H, H→jj events at 240 GeV, and compared with the domain-specific models ParticleNet and Particle Transformer. The authors report that BBT-Neutron reaches comparable performance at 10 million training events and exhibits a sharp scaling transition ("emergence") in flavor-tagging efficiency and charge-flip rate as a function of training dataset size, which they interpret as evidence for a generalist-model phase transition and a step toward foundation models for particle physics.

Significance. If the main claims are supported, the result is of interest to both the ML-for-science and high-energy physics communities: it would demonstrate that a generic byte-level transformer with no physics-specific inductive biases can match specialized jet-tagging architectures, and that binary tokenization is a viable alternative to BPE for numerical data. The evaluation is a standard supervised train/test split against external baselines, and the code is open-sourced, so the central performance question is not circular. However, the current evidence is primarily visual and the scaling interpretation is confounded, so the significance is conditional on additional quantitative and controlled experiments.

major comments (4)
  1. [4.1.3 and Figs. 4-5] Section 4.1.3 fixes batch size at 512 and training at 30 epochs for all dataset sizes. Consequently, the total number of optimizer steps grows roughly linearly with dataset size: about 30 steps for 100 events and about 5.9×10^5 steps for 10^7 events. The x-axis of the scaling plots (Figs. 4 and 5) therefore confounds dataset size with total compute/optimizer steps. The sharp jump in BBT-Neutron's curves may be an undertraining artifact for small dataset sizes, and no learning curves or convergence checks are reported for any dataset size. This is load-bearing for the emergence interpretation in Section 2. The authors should hold compute fixed, train to convergence, or otherwise separate the data-size effect from the optimizer-step effect, and report loss curves.
  2. [Section 2 and Figs. 2-5] The central claims of comparable performance and of a phase-transition-like jump rest on visual inspection of confusion matrices and scaling curves. The manuscript gives no numeric values for the flavor-tagging efficiencies or charge-flip rates, no error bars, no repeated-seed variability, and no statistical test or fitted scaling law for the proposed critical data thresholds. Phrases such as "strongly indicates the emergence of a generalist architecture" go beyond what the plotted curves alone demonstrate. Add tables with point estimates and uncertainties, and a formal test (e.g., confidence intervals or a comparison of fitted models) for the differential scaling behavior between BBT-Neutron and the specialized baselines.
  3. [4.2 and Figs. 2-5] To establish "comparable performance" to ParticleNet and Particle Transformer, the baselines need to be trained on the same data with a matched or at least reported training budget. The manuscript does not provide the training details, hyperparameters, parameter counts, or compute used for PN and ParT. Without this information, and given the fixed-budget issue above, the comparison in Figures 2-5 is not on an equal footing. Please report the baseline training protocol and, ideally, compute-matched scaling curves.
  4. [Section 5] The Conclusion states that "BBT-Neutron achieves state-of-the-art performance in Jet Origin Identification," but the Abstract and Results claim only that its performance is comparable to that of two specific models. Because no quantitative comparison with a broader set of methods is shown, this sentence overstates the evidence. Please either provide quantitative evidence for state-of-the-art status or soften the claim to comparability with the evaluated baselines.
minor comments (5)
  1. [Keywords and Abstract] Correct the typo "Jet Original Identification" to "Jet Origin Identification," and replace "spacial" with "spatial" in the Abstract and Section 3.
  2. [4.1.2] The tokenization section calls the representation "octal (base-8) byte sequences," but the worked examples show decimal byte arrays (e.g., [49, 50, 51, 52, 53]) and mention hexadecimal as an option. Please define the exact byte encoding used in the experiments, including endianness and width for the native numerical conversion, since this is the method's main contribution.
  3. [Abstract and 4.1.3] The abstract says BBT-Neutron uses binary tokenization "to facilitate pretraining on a mixture of textual and large-scale numerical experimental data," but the experiments are end-to-end supervised training on JoI only and no pretraining on a multimodal mixture is reported. Please clarify whether any pretraining was performed or whether the current study only demonstrates the architecture's supervised use.
  4. [References] Reference [51] is incomplete: "Vaswani, A.: Attention is all you need" lacks co-authors and publication venue; cite the full NeurIPS 2017 paper or a version of record.
  5. [Figures 4-5] The panels in Figures 4 and 5 would be easier to read with explicit labels for the BBT-Neutron, ParticleNet, and Particle Transformer curves, since the caption text is the only guide in the grayscale version.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central JoI performance and scaling claims are evaluated independently against external baselines and are not derived from fitted parameters or self-citations.

full rationale

The paper's central claims are (1) that BBT-Neutron achieves comparable performance to state-of-the-art task-specific JoI models, and (2) that its performance exhibits a sharp scaling transition with increasing training data. Neither claim is circular. The performance claim rests on a standard supervised evaluation with held-out test data and direct comparison against ParticleNet and Particle Transformer, which are external, independently published baselines cited as references [46] and [50]. The scaling claim is presented as an empirical observation of measured accuracy versus dataset size, not as a prediction derived from a fitted scaling law, so there is no fitted-input-called-prediction pattern. No load-bearing argument reduces to a self-citation: the authors do cite prior work by co-author M. Ruan (CEPC CDR [29]) and the Particle Transformer paper by Qu, Li, and Qian [46], but these citations provide the dataset, detector simulation, and baseline architecture rather than justifying the paper's own conclusions. No uniqueness theorem, ansatz, or definitional equivalence is imported from the authors' own prior work. The well-known confound noted in the skeptical reading, that fixed 30 epochs and batch size 512 make total optimizer steps grow with dataset size, is a genuine experimental-design concern about whether the small-dataset runs are undertrained, but it is not a circularity: the reported numbers are still measurements, not quantities forced by construction. Accordingly, the circularity score is 0.

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

No new physical entities are introduced. The ledger captures standard training hyperparameters that are chosen by hand, plus implicit assumptions about data and baseline fairness that the paper does not verify.

free parameters (5)
  • initial learning rate = 1e-4
    Set for training with cosine schedule; not tuned or justified across dataset sizes.
  • batch size = 512
    Fixed for all runs; no sensitivity study.
  • training epochs = 30
    Fixed budget for every dataset size; central to the scaling comparison and possible under-training of small datasets.
  • model parameter count = 160 million
    Only one scale is tested; no model-scaling study.
  • patch size = not stated for experiments
    Section 4.1.2 gives Patch Size=16 as an example, but the actual value used for JoI is not reported.
assumptions (4)
  • domain assumption CEPC simulated events at sqrt(s)=240 GeV with the baseline detector are a valid testbed for JoI.
    The paper uses simulated nu-nu H, H->jj events and assumes they represent the physics task; real detector effects may differ.
  • domain assumption The 19 particle-level features listed in Table 1 are sufficient for jet origin identification.
    All models receive the same features; if important information is missing, the comparison could be biased.
  • domain assumption Baseline models ParticleNet and Particle Transformer are trained on the same data and setup.
    No training details for baselines are provided, so the fairness of the comparison is assumed.
  • ad hoc to paper A fixed 30-epoch schedule is sufficient for convergence at all dataset sizes.
    This is the key assumption behind the claimed emergence transition; it is not checked.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Scaling Particle Collision Data Analysis." pith.science (2026). https://pith.science/paper/JFHUPYUX

@misc{pith2026241200129,
  author       = {Pith},
  title        = {Pith review of: Scaling Particle Collision Data Analysis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JFHUPYUX}},
  note         = {Machine review of arXiv:2412.00129}
}
read the original abstract

For decades, researchers have developed task-specific models to address scientific challenges across diverse disciplines. Recently, large language models (LLMs) have shown enormous capabilities in handling general tasks; however, these models encounter difficulties in addressing real-world scientific problems, particularly in domains involving large-scale numerical data analysis, such as experimental high energy physics. This limitation is primarily due to BPE tokenization's inefficacy with numerical data. In this paper, we propose a task-agnostic architecture, BBT-Neutron, which employs a binary tokenization method to facilitate pretraining on a mixture of textual and large-scale numerical experimental data. We demonstrate the application of BBT-Neutron to Jet Origin Identification (JoI), a critical categorization challenge in high-energy physics that distinguishes jets originating from various quarks or gluons. Our results indicate that BBT-Neutron achieves comparable performance to state-of-the-art task-specific JoI models. Furthermore, we examine the scaling behavior of BBT-Neutron's performance with increasing data volume, suggesting the potential for BBT-Neutron to serve as a foundational model for particle physics data analysis, with possible extensions to a broad spectrum of scientific computing applications for Big Science experiments, industrial manufacturing and spacial computing. The project code is available at https://github.com/supersymmetry-technologies/bbt-neutron.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 19 canonical work pages

  1. [1]

    The C Users Journal archive 12, 23–38 (1994)

    Gage, P.: A new algorithm for data compression. The C Users Journal archive 12, 23–38 (1994)

    work page 1994

  2. [2]

    arXiv preprint arXiv:2108.07258v2 (2021)

    Bommasani, R., Hudson, D.A., et al.: On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258v2 (2021)

    arXiv 2021

  3. [4]

    arXiv preprint arXiv:2001.08361v1 (2020) 18

    Henighan, T., Brown, T.B., et al.: Scaling laws for neural language models. arXiv preprint arXiv:2001.08361v1 (2020) 18

    arXiv 2020

  4. [5]

    OpenAI blog 1(8), 9 (2019)

    Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., Sutskever, I., et al.: Language models are unsupervised multitask learners. OpenAI blog 1(8), 9 (2019)

    2019

  5. [6]

    arXiv preprint arXiv:2005.14165 (2020)

    Brown, T.B.: Language models are few-shot learners. arXiv preprint arXiv:2005.14165 (2020)

    arXiv 2020

  6. [7]

    arXiv preprint arXiv:2203.15556 (2022)

    Hoffmann, J., Borgeaud, S., Mensch, A., Buchatskaya, E., Cai, T., Rutherford, E., Casas, D.d.L., Hendricks, L.A., Welbl, J., Clark, A., et al.: Training compute- optimal large language models. arXiv preprint arXiv:2203.15556 (2022)

    arXiv 2022

  7. [8]

    arXiv preprint arXiv:2402.17177 (2024)

    Liu, Y., Zhang, K., Li, Y., Yan, Z., Gao, C., Chen, R., Yuan, Z., Huang, Y., Sun, H., Gao, J., et al.: Sora: A review on background, technology, limitations, and opportunities of large vision models. arXiv preprint arXiv:2402.17177 (2024)

    arXiv 2024

  8. [9]

    arXiv preprint arXiv:2309.03241 (2023)

    Yang, Z., Ding, M., Lv, Q., Jiang, Z., He, Z., Guo, Y., Bai, J., Tang, J.: Gpt can solve mathematical problems without a calculator. arXiv preprint arXiv:2309.03241 (2023)

    arXiv 2023

Show all 50 references
  1. [10]

    https://arxiv.org/abs/2106.04560

    Zhai, X., Kolesnikov, A., Houlsby, N., Beyer, L.: Scaling Vision Transformers (2022). https://arxiv.org/abs/2106.04560

    2022 arXiv

  2. [11]

    https: //arxiv.org/abs/2212.09748

    Peebles, W., Xie, S.: Scalable Diffusion Models with Transformers (2023). https: //arxiv.org/abs/2212.09748

    2023 arXiv

  3. [12]

    https://arxiv.org/abs/2010

    Henighan, T., Kaplan, J., Katz, M., Chen, M., Hesse, C., Jackson, J., Jun, H., Brown, T.B., Dhariwal, P., Gray, S., Hallacy, C., Mann, B., Radford, A., Ramesh, A., Ryder, N., Ziegler, D.M., Schulman, J., Amodei, D., McCandlish, S.: Scaling Laws for Autoregressive Generative Mo...

    2020

  4. [13]

    https://arxiv.org/abs/2402.02054

    Liu, J., Mao, H., Chen, Z., Zhao, T., Shah, N., Tang, J.: Neural Scaling Laws on Graphs (2024). https://arxiv.org/abs/2402.02054

    2024 arXiv

  5. [14]

    https://arxiv.org/abs/2301.13442

    Hilton, J., Tang, J., Schulman, J.: Scaling laws for single-agent reinforcement learning (2023). https://arxiv.org/abs/2301.13442

    2023 arXiv

  6. [15]

    Nature communications 12(1), 5721 (2021)

    Gupta, A., Savarese, S., Ganguli, S., Fei-Fei, L.: Embodied intelligence via learning and evolution. Nature communications 12(1), 5721 (2021)

    2021

  7. [16]

    arXiv preprint arXiv:2403.00504 (2024)

    Garrido, Q., Assran, M., Ballas, N., Bardes, A., Najman, L., LeCun, Y.: Learning and leveraging world models in visual representation learning. arXiv preprint arXiv:2403.00504 (2024)

    2024 arXiv

  8. [17]

    arXiv preprint arXiv:2407.21783 (2024) 19

    Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., Yang, A., Fan, A., et al.: The llama 3 herd of models. arXiv preprint arXiv:2407.21783 (2024) 19

    2024 arXiv

  9. [18]

    Yuan, Z., Yuan, H., Tan, C., Wang, W., Huang, S.: How well do large language models perform in arithmetic tasks? arXiv preprint arXiv:2304.02015 (2023)

    2023 arXiv

  10. [19]

    ArXiv abs/2405.17399 (2024)

    McLeish, S., Bansal, A., Stein, A., Jain, N., Kirchenbauer, J., Bartoldson, B., Kailkhura, B., Bhatele, A., Geiping, J., Schwarzschild, A., Goldstein, T.: Trans- formers can do arithmetic with the right embeddings. ArXiv abs/2405.17399 (2024)

    2024 arXiv

  11. [20]

    Nature, 1–3 (2024)

    Abramson, J., Adler, J., Dunger, J., Evans, R., Green, T., Pritzel, A., Ron- neberger, O., Willmore, L., Ballard, A.J., Bambrick, J., et al.: Accurate structure prediction of biomolecular interactions with alphafold 3. Nature, 1–3 (2024)

    2024

  12. [21]

    nature 596(7873), 583–589 (2021)

    Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., ˇZ ´ ıdek, A., Potapenko, A.,et al.: Highly accurate protein structure prediction with alphafold. nature 596(7873), 583–589 (2021)

    2021

  13. [22]

    Incentivize: Scale-first view of Large Language Models

    Chung, H.W.: Don’t teach. Incentivize: Scale-first view of Large Language Models. PPT https://www.csail.mit.edu/event/ ei-seminar-hyung-won-chung-openai-dont-teach-incentivize-scale-first-view-large-language (2024)

    2024

  14. [23]

    arXiv preprint arXiv:2211.09085 (2022)

    Taylor, R., Kardas, M., Cucurull, G., Scialom, T., Hartshorn, A., Saravia, E., Poulton, A., Kerkez, V., Stojnic, R.: Galactica: A large language model for science. arXiv preprint arXiv:2211.09085 (2022)

    2022 arXiv

  15. [24]

    arXiv preprint arXiv:2308.13565 (2023)

    Xie, T., Wan, Y., Huang, W., Yin, Z., Liu, Y., Wang, S., Linghu, Q., Kit, C., Grazian, C., Zhang, W., et al.: Darwin series: Domain specific large language models for natural science. arXiv preprint arXiv:2308.13565 (2023)

    2023 arXiv

  16. [25]

    arXiv preprint arXiv:2403.03154 (2024)

    Pan, H., Mudur, N., Taranto, W., Tikhanovskaya, M., Venugopalan, S., Bahri, Y., Brenner, M.P., Kim, E.-A.: Quantum many-body physics calculations with large language models. arXiv preprint arXiv:2403.03154 (2024)

    2024 arXiv

  17. [26]

    Nature 625(7995), 468–475 (2024)

    Romera-Paredes, B., Barekatain, M., Novikov, A., Balog, M., Kumar, M.P., Dupont, E., Ruiz, F.J., Ellenberg, J.S., Wang, P., Fawzi, O., et al.: Mathematical discoveries from program search with large language models. Nature 625(7995), 468–475 (2024)

    2024

  18. [27]

    arXiv preprint arXiv:2403.07187 (2024)

    Shen, J., Marwah, T., Talwalkar, A.: Ups: Towards foundation models for pde solving via cross-modal adaptation. arXiv preprint arXiv:2403.07187 (2024)

    2024 arXiv

  19. [28]

    arXiv preprint arXiv:2404.07544 (2024)

    Vacareanu, R., Negru, V.-A., Suciu, V., Surdeanu, M.: From words to numbers: Your large language model is secretly a capable regressor when given in-context examples. arXiv preprint arXiv:2404.07544 (2024)

    2024 arXiv

  20. [29]

    https://arxiv.org/abs/1811.10545 20

    The CEPC Study Group: CEPC Conceptual Design Report: Volume 2 - Physics & Detector (2018). https://arxiv.org/abs/1811.10545 20

    2018 arXiv

  21. [30]

    The European Physical Journal C-Particles and Fields 12(2), 183– 207 (2000)

    collaboration, A., Barate, R.: Study of fermion pair production in collisions at 130–183 gev. The European Physical Journal C-Particles and Fields 12(2), 183– 207 (2000)

    2000

  22. [31]

    Physical review letters 80(4), 660 (1998)

    Abe, K., Akagi, T., Allen, N., Ash, W., Aston, D., Baird, K., Baltay, C., Band, H., Barakat, M., Baranko, G., et al.: Measurement of r b using a vertex mass tag. Physical review letters 80(4), 660 (1998)

    1998

  23. [32]

    Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 417(2-3), 384–390 (1998)

    Borisov, G.: Combined b-tagging. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 417(2-3), 384–390 (1998)

    1998

  24. [33]

    arXiv preprint hep-ex/0311003 (2003)

    Abdallah, J., Collaboration, D., et al.: b-tagging in delphi at lep. arXiv preprint hep-ex/0311003 (2003)

    2003 arXiv

  25. [34]

    Journal of Instrumentation 13(5), 05011 (2018)

    collaboration, C., et al.: Identification of heavy-flavour jets with the cms detector in pp collisions at 13 tev. Journal of Instrumentation 13(5), 05011 (2018)

    2018

  26. [35]

    collaboration, C., et al.: Identification of b quark jets at the cms experiment in the lhc run 2 (2016)

    2016

  27. [36]

    arXiv preprint arXiv:1711.09059 (2017)

    Egan, S., Fedorko, W., Lister, A., Pearkes, J., Gay, C.: Long short-term memory (lstm) networks with jet constituents for boosted top tagging at the lhc. arXiv preprint arXiv:1711.09059 (2017)

    2017 arXiv

  28. [37]

    Physical Review D 94(11), 112002 (2016)

    Guest, D., Collado, J., Baldi, P., Hsu, S.-C., Urban, G., Whiteson, D.: Jet flavor classification in high-energy physics with deep neural networks. Physical Review D 94(11), 112002 (2016)

    2016

  29. [38]

    Journal of High Energy Physics 2015(2), 1–16 (2015)

    Cogan, J., Kagan, M., Strauss, E., Schwarztman, A.: Jet-images: computer vision inspired techniques for jet tagging. Journal of High Energy Physics 2015(2), 1–16 (2015)

    2015

  30. [39]

    Journal of High Energy Physics 2016(7), 1–32 (2016)

    Oliveira, L., Kagan, M., Mackey, L., Nachman, B., Schwartzman, A.: Jet- images—deep learning edition. Journal of High Energy Physics 2016(7), 1–32 (2016)

    2016

  31. [40]

    https://arxiv.org/abs/2206.07682

    Wei, J., Tay, Y., Bommasani, R., Raffel, C., Zoph, B., Borgeaud, S., Yogatama, D., Bosma, M., Zhou, D., Metzler, D., Chi, E.H., Hashimoto, T., Vinyals, O., Liang, P., Dean, J., Fedus, W.: Emergent Abilities of Large Language Models (2022). https://arxiv.org/abs/2206.07682

    2022 arXiv

  32. [41]

    https://arxiv.org/abs/2410

    Lin, F., Hu, Y., Sheng, P., Wen, C., You, J., Gao, Y.: Data Scaling Laws in Imitation Learning for Robotic Manipulation (2024). https://arxiv.org/abs/2410. 18647

    2024

  33. [42]

    https://arxiv.org/abs/2402

    Wu, S., Tan, X., Wang, Z., Wang, R., Li, X., Sun, M.: Beyond Language Models: 21 Byte Models are Digital World Simulators (2024). https://arxiv.org/abs/2402. 19155

    2024

  34. [43]

    https://arxiv.org/abs/2305.07185

    Yu, L., Simig, D., Flaherty, C., Aghajanyan, A., Zettlemoyer, L., Lewis, M.: MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers (2023). https://arxiv.org/abs/2305.07185

    2023 arXiv

  35. [44]

    https://arxiv.org/abs/2404.14408

    Slagle, K.: SpaceByte: Towards Deleting Tokenization from Large Language Modeling (2024). https://arxiv.org/abs/2404.14408

    2024 arXiv

  36. [45]

    Neurocomputing 568, 127063 (2024) https://doi.org/10.1016/j.neucom.2023.127063

    Su, J., Ahmed, M., Lu, Y., Pan, S., Bo, W., Liu, Y.: Roformer: Enhanced trans- former with rotary position embedding. Neurocomputing 568, 127063 (2024) https://doi.org/10.1016/j.neucom.2023.127063

    2024

  37. [46]

    Qu, H., Li, C., Qian, S.: Particle Transformer for Jet Tagging (2022) arXiv:2202.03772 [hep-ph]

    2022 arXiv

  38. [47]

    AI Open 1, 57–81 (2020) https://doi.org/10.1016/j.aiopen.2021.01.001

    Zhou, J., Cui, G., Hu, S., Zhang, Z., Yang, C., Liu, Z., Wang, L., Li, C., Sun, M.: Graph neural networks: A review of methods and applications. AI Open 1, 57–81 (2020) https://doi.org/10.1016/j.aiopen.2021.01.001

    2020 doi

  39. [48]

    Nature Communications 14(1), 35 (2023) https://doi.org/10

    Bannigan, P., Bao, Z., Hickman, R.J., Aldeghi, M., H¨ ase, F., Aspuru-Guzik, A., Allen, C.: Machine learning models to accelerate the design of polymeric long- acting injectables. Nature Communications 14(1), 35 (2023) https://doi.org/10. 1038/s41467-022-35343-w

    2023

  40. [49]

    https://arxiv.org/abs/1508.07909

    Sennrich, R., Haddow, B., Birch, A.: Neural Machine Translation of Rare Words with Subword Units (2016). https://arxiv.org/abs/1508.07909

    2016 arXiv

  41. [50]

    Qu, H., Gouskos, L.: ParticleNet: jet tagging via particle clouds. Phys. Rev. D 101(5), 056019 (2020) https://doi.org/10.1103/PhysRevD.101.056019 arXiv:1902.08570 [hep-ph]

    2020 arXiv

  42. [51]

    Advances in Neural Information Processing Systems (2017) 22

    Vaswani, A.: Attention is all you need. Advances in Neural Information Processing Systems (2017) 22

    2017

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.