Pith. sign in

REVIEW 3 major objections 6 minor 2 cited by

Limit Order Book Event Stream Prediction with Diffusion Model

T0 review · 3 major / 6 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 LOBDIF models the next limit-order-book event's time and type as the endpoint of a Gaussian diffusion chain conditioned on history, and reports lower error and higher accuracy than point-process baselines on six datasets.

desk verdict New diffusion-based approach to LOB event-stream prediction is worth engaging, but the reported Wilcoxon p-values cannot come from five paired runs and the discrete event-type decoding rule is missing; major revision needed before the significance claim is credible. read the letter →

arxiv 2412.09631 v1 pith:3SSTKHMD submitted 2024-11-27 q-fin.ST

Zetao Zheng , Guoan Li , Deqiang Ouyang , Decui Liang , Jie Shao This is my paper · ORCID
classification q-fin.ST
keywords limitorderbookeventstreampredictiondiffusionmodelmarkedpointprocessHawkestime-eventjointdistributionskip-stepsamplingdenoisingnetwork
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

The paper proposes LOBDIF, a diffusion model for limit order book event stream prediction. It treats each event as a pair of arrival time and event type, and learns the joint distribution of the next event by decomposing it into a Markov chain of Gaussian denoising steps conditioned on the encoded history of previous events. The authors claim this captures time-event interdependence better than stochastic and neural point processes, and report lower next-event-time MAE and higher event-type accuracy than Hawkes, LSTM, SAHP, CT-LSTM, and PCT-LSTM baselines on six datasets from three traded assets. A skip-step sampling strategy is claimed to make inference about an order of magnitude faster with negligible accuracy loss, making the approach practical for high-frequency LOB prediction.

What carries the argument

The load-bearing object is the conditional diffusion chain over the event tuple. The forward process in Eq. (4) adds Gaussian noise to time and event type with a shared variance schedule, and the reverse process in Eq. (5) is conditioned on the historical context $h_{i-1}$ produced by a self-attention encoder over time, event, and joint embeddings. A dedicated denoising network computes time attention and event attention at each step to predict the noise, and skip-step sampling, a DDIM-style reparameterization, lets inference jump from step $k$ to step $s < k$. The whole mechanism converts the intractable joint time-event distribution into a sequence of tractable Gaussian transitions.

What would settle it

A direct check would be to change only the decoding of the event-type coordinate and re-run Table III on MSFT1 with nearest-integer rounding, with argmax over the learned event embedding, and with a small learned classifier head on the denoised output. If any of these reproduces the reported accuracy only through a rule the paper does not state, or if the accuracy drops below PCT-LSTM under a natural rule, the event-type claim is not carried by the generative model.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a diffusion model can serve as a replacement for intensity-based point processes in LOB event stream prediction. Instead of assuming a Poisson or Hawkes intensity that governs event arrivals, LOBDIF adds Gaussian noise to both coordinates of each event $(t_i, e_i)$ over $K$ steps, then learns a reverse process that removes the noise while attending to the historical context $h_{i-1}$. Because each reverse step is a Gaussian transition, sampling is closed-form; because the target is the full time-event distribution, the model can predict time and type jointly. The reported experiments show LOBDIF beating the strongest baselines on most datasets, with Wilcoxon signed-rank p-values below 0.05, and the case-study visualization shows the denoised distribution converging toward the empirical time-event scatter.

Load-bearing premise

The forward diffusion adds Gaussian noise to the discrete event-type label as if it were a continuous number, and the paper never says how the denoised continuous value is turned back into an event class, so the reported event-type accuracy rests on an unstated conversion step.

Editorial extensions

If this is right

  • If LOBDIF's central claim is right, next-event prediction in LOB systems no longer needs an intensity function; the joint distribution over time and type can be sampled in closed form through Gaussian steps.
  • The skip-step sampling result means the same trained model can trade speed against latency by picking a step size at inference time, with no retraining.
  • Because the diffusion decomposition replaces a single global intensity model with many local Gaussian transitions, the approach should track non-stationary order-flow regimes better than Hawkes-style models.
  • On the reported datasets, removing time encoding hurts more than removing event encoding, pointing to event times as the harder channel to model.

Reading between the lines

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

  • Editorial extension: the paper's event-type channel is diffused as a continuous scalar without a stated rounding rule; a discrete-diffusion variant over the one-hot label would make the type channel fully grounded and is the natural next test.
  • Editorial extension: because the shared Gaussian noise schedule treats time and type on the same scale, rescaling time by its empirical spread before diffusion could change the MAE results without altering the architecture.
  • Editorial extension: the skip-step sampler is largely agnostic to the denoising network, so the same conditioning scheme could be applied to other irregular event streams such as clickstreams or social-media activity, where closed-form sampling also matters.
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

3 major / 6 minor

Summary. The paper proposes LOBDIF, a conditional diffusion model for predicting the next event in a limit order book event stream: given the last L events (time and type), it learns a joint time-event distribution by adding Gaussian noise to both coordinates and training a history-conditioned denoising network, then generates predictions by reverse denoising with a skip-step strategy. Experiments on six datasets from three assets compare LOBDIF with Hawkes, LSTM, CT-LSTM, SAHP, and PCT-LSTM, reporting lower MAE and higher accuracy, and use paired Wilcoxon tests to claim statistical significance.

Significance. The idea of applying diffusion models to marked point-process-style event streams is timely and, if the empirical claims hold, would provide a useful alternative to neural Hawkes processes, with reported millisecond-level per-event inference times. Strengths of the paper are the use of real order-book data from two databases, the availability of code, and a clear ablation structure. The contribution is incremental relative to standard conditional DDPMs, and the main empirical conclusion currently rests on a statistically impossible significance test; the event-type decoding rule is also missing. With those issues corrected, the paper could be acceptable as an application-oriented contribution.

major comments (3)
  1. [Section VI-B, Table III] The paired Wilcoxon signed-rank test described in Section VI-B is based on '5 pairs of experimental results,' i.e., five runs of LOBDIF and five runs of the strongest baseline. For n=5 paired observations, the smallest attainable two-sided Wilcoxon signed-rank p-value is 2/2^5 = 0.0625, and the smallest one-sided p-value is 1/32 = 0.03125. Every p-value in Table III (1.52e-3 to 7.63e-3) is therefore impossible under the stated protocol. The sentence 'all p-values are below 0.05, confirming that the experimental results are statistically significant' is unsupported. Please disclose the actual unit of pairing (for example, per-event paired predictions, which would yield a much larger n) and recompute all p-values, or remove the significance language and treat Table III as point estimates.
  2. [Section IV-A, Eq. (4); Section VI-B] Equation (4) adds Gaussian noise to the event-type coordinate e_i exactly as if it were a continuous scalar, although e_i is a categorical label in {0,...,3} for the LOBSTER datasets and has three classes for the CSMAR datasets. The reverse process in Section IV-B and Algorithm 2 operates on the same continuous coordinate and returns a continuous x0_i. The paper never specifies how the denoised continuous e0_i is mapped to a class label for the reported Acc. metric, and it does not include a rounding rule, one-hot reconstruction, or any discrete-diffusion correction. Without this decoding rule, the event-type accuracy numbers in Table III are not attributable to the generative model. Please specify the decoding rule or replace the categorical diffusion formulation with one that is defined for discrete event types.
  3. [Section III-B, Eqs. (1)-(2); Eq. (11)] The forward-process equations are internally inconsistent. Equation (1) writes x_k = sqrt(1-β_k) x_{k-1} + β_k ε_k and then x_k = sqrt(α_k) x_0 + (1-α_k) ε_k, while Eq. (3) and the training loss in Eq. (11) use sqrt(1-α_k) ε_k. The standard DDPM recursion requires sqrt(β_k) ε and sqrt(1-α_k) ε; as written, Eq. (1) does not imply the stated closed-form marginal. Equation (2) also contains ambiguous fractions and missing square roots in the posterior mean. Please correct these equations so that the forward process, the training objective, and the skip-step sampling derivation in Eq. (12) are mutually consistent.
minor comments (6)
  1. [Section I; Table I] The term 'closed-form sampling' and the Table I entry 'Sampling without any approximation' overstate the method: the reverse process of Eq. (2) is iterative and stochastic, and the skip-step strategy in Section V-B reduces but does not eliminate iterative denoising. Consider rewording to 'efficient ancestral sampling' or similar.
  2. [Section VI-A.4] The text states that next-event type accuracy is assessed both when the event time is known and unknown, but Table III reports only a single Acc. column. Please clarify which protocol is actually used.
  3. [Table III] The p-value row contains stray spaces (e.g., '4 .27e-3') and should be reformatted for readability.
  4. [Section VI-C.2] The text says 'Due to space limitations, we present results for only three datasets,' but Table V contains results for all six datasets. The statement should be corrected.
  5. [Section VI-A.1; Abstract] 'China Telecommute' should be 'China Telecom,' and 'Hawke stochastic process' should be 'Hawkes stochastic process.' These typos appear in the abstract and body text.
  6. [Figure 2] Figure 2 is too small and low-resolution to read the framework details, especially the forward/reverse and conditioning arrows. A larger, clearer figure is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: LOBDIF is trained on held-out training splits, evaluated on fresh test events, and its diffusion components come from external standard DDPM machinery; reported statistical-significance issues are correctness risks, not circularity.

full rationale

The derivation chain is self-contained against external benchmarks. The forward process in Eq. (4) adds Gaussian noise to time and event type values, and the training objective in Eq. (11) is the standard DDPM noise-prediction loss; no fitted parameter is renamed as a prediction. The reverse process in Eq. (5) conditions on an encoded history context, and the skip-step sampling in Eq. (12) follows the standard DDIM reparameterization from external prior work. The central claim is an empirical comparison against several baselines on held-out test events, trained on separate splits, so the predictions are not fitted to the test data by construction. There are no load-bearing self-citations: the diffusion and attention machinery cited is external standard work, and no uniqueness theorem or prior result by the same authors is used to force the model choice. The reported p-values in Table III appear impossible for five paired runs, and the paper does not describe how continuous denoised event values are decoded into discrete class labels for accuracy, but these are validity and reproducibility concerns rather than circular reductions. The paper's derivation does not reduce to its own inputs, so no circular step is identified.

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

The central claim rests on standard diffusion model machinery plus a set of modeling choices: a 50-event Markov window, a conditional independence factorization between time and event type, and the treatment of categorical event types as continuous under Gaussian noise. The last assumption is the least supported and is load-bearing for the reported event-type accuracy. Hyperparameters such as L, M, K, tau, learning rate, and epochs are manually selected; the main experiments do not report the exact K and tau used.

free parameters (5)
  • history length L = 50
    Set in Section VI-A without reporting a sensitivity analysis; all substreams use L=50.
  • embedding dimension M = 64
    Section VI-A sets time and event encoding dimensions to 64; Section VI-D3 shows this is a manually chosen trade-off.
  • diffusion steps K = Not explicitly reported for main experiments; case study uses 100
    Hyperparameter analysis in Section VI-D1 tests K={50,100,400,800,1000}, but the value used for Table III results is not stated.
  • skip-step interval tau = Not reported for main results
    Ablation in Figure 4 tests tau={5,10,20,50}, but the tau used to produce the main comparison table is unspecified.
  • learning rate and training epochs = 2e-3 and 200 epochs
    Section VI-A lists these values as chosen settings; Section VI-D2 shows 200 epochs is a trade-off.
assumptions (4)
  • domain assumption The next event depends only on the previous L=50 events.
    Section III-A defines L-sized substreams and the task as predicting event L+1 from the previous L events; sensitivity to L is not examined.
  • domain assumption Event time and event type are conditionally independent in each reverse denoising step given history and step.
    Eq. (5) factorizes p_theta(x_{k-1}|x_k,h,k) into a time term and an event type term, and the denoising network predicts time noise and event noise separately.
  • ad hoc to paper A categorical event type can be treated as a real-valued quantity under additive Gaussian noise.
    Eq. (4) applies the same Gaussian forward process to e_i as to t_i, despite e_i being one of three or four discrete classes; no discrete diffusion or reconstruction likelihood is provided.
  • standard math Standard DDPM and DDIM theory is valid for the time component and for the reverse process.
    Eqs. (1)-(3) and Algorithm 2 rely on the standard Gaussian diffusion derivation from Ho et al. and Song et al. without restating the proofs.
invented entities (1)
  • None
    purpose: No new physical or mathematical entities are introduced.
    The paper introduces new neural network modules (time-event encoder, time/event attention denoiser, skip-step sampling) but these are implementation components, not new postulated entities such as particles, mediators, forces, or conserved quantities.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Limit Order Book Event Stream Prediction with Diffusion Model." pith.science (2026). https://pith.science/paper/3SSTKHMD

@misc{pith2026241209631,
  author       = {Pith},
  title        = {Pith review of: Limit Order Book Event Stream Prediction with Diffusion Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3SSTKHMD}},
  note         = {Machine review of arXiv:2412.09631}
}
read the original abstract

Limit order book (LOB) is a dynamic, event-driven system that records real-time market demand and supply for a financial asset in a stream flow. Event stream prediction in LOB refers to forecasting both the timing and the type of events. The challenge lies in modeling the time-event distribution to capture the interdependence between time and event type, which has traditionally relied on stochastic point processes. However, modeling complex market dynamics using stochastic processes, e.g., Hawke stochastic process, can be simplistic and struggle to capture the evolution of market dynamics. In this study, we present LOBDIF (LOB event stream prediction with diffusion model), which offers a new paradigm for event stream prediction within the LOB system. LOBDIF learns the complex time-event distribution by leveraging a diffusion model, which decomposes the time-event distribution into sequential steps, with each step represented by a Gaussian distribution. Additionally, we propose a denoising network and a skip-step sampling strategy. The former facilitates effective learning of time-event interdependence, while the latter accelerates the sampling process during inference. By introducing a diffusion model, our approach breaks away from traditional modeling paradigms, offering novel insights and providing an effective and efficient solution for learning the time-event distribution in order streams within the LOB system. Extensive experiments using real-world data from the limit order books of three widely traded assets confirm that LOBDIF significantly outperforms current state-of-the-art methods.

Figures

Figures reproduced from arXiv: 2412.09631 by the authors.

Figure 1
Figure 1. An LOB with four price levels evolving over time. The event at [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. The overall framework of our proposed method. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Average running time comparison for testing dataset execution. [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Average running time comparison for testing dataset execution. [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Analysis of the proposed model’s performance with different hyperparameters. [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: Visualization of the time-event distribution. The first five columns (in blue) illustrate different stages of the denoising process, while the final column [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. FlowLOB: Efficient and Controllable Limit Order Book Generation with Flow Matching

    cs.LG 2026-08 conditional novelty 6.0 of 10

    Flow matching with a tick-relative LOB representation and transformer backbone generates realistic, controllable, and cross-instrument limit order book states at low sampling cost on HKEX data.

  2. Exploring Microstructural Dynamics in Cryptocurrency Limit Order Books: Better Inputs Matter More Than Stacking Another Hidden Layer

    cs.LG 2025-06 reject novelty 4.0 of 10

    On one day of BTC/USDT order book data, Savitzky-Golay smoothing and feature choice helped simple models match or beat deeper neural networks.

Reference graph

Works this paper leans on

39 extracted references · 35 canonical work pages · cited by 2 Pith papers

  1. [1]

    A stochastic model for order book dynamics,

    R. Cont, S. Stoikov, and R. Talreja, “A stochastic model for order book dynamics,” Oper. Res., vol. 58, no. 3, pp. 549–563, 2010

    work page 2010

  2. [2]

    Long-time behavior of a hawkes process- based limit order book,

    F. Abergel and A. Jedidi, “Long-time behavior of a hawkes process- based limit order book,” SIAM J. Financial Math. , vol. 6, no. 1, pp. 1026–1043, 2015

    work page 2015

  3. [3]

    The neural hawkes process: A neurally self- modulating multivariate point process,

    H. Mei and J. Eisner, “The neural hawkes process: A neurally self- modulating multivariate point process,” in Advances in Neural Informa- tion Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017 , 2017, pp. 6754–6764

    work page 2017

  4. [4]

    Self-attentive hawkes process,

    Q. Zhang, A. Lipani, ¨O. Kirnap, and E. Yilmaz, “Self-attentive hawkes process,” in Proceedings of the 37th International Conference on Ma- chine Learning, ICML 2020 , 2020, pp. 11 183–11 193

    work page 2020

  5. [5]

    Deep Reinforcement Learning for Market Making Under a Hawkes Process-Based Limit Order Book Model

    B. Gasperov and Z. Kostanjcar, “Deep reinforcement learning for market making under a hawkes process-based limit order book model,” CoRR, vol. abs/2207.09951, 2022

    work page Pith review arXiv 2022

  6. [6]

    State dependent parallel neural hawkes process for limit order book event stream prediction and simulation,

    Z. Shi and J. Cartlidge, “State dependent parallel neural hawkes process for limit order book event stream prediction and simulation,” in KDD ’22: The 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022, pp. 1607–1615

    work page 2022

  7. [7]

    Neural marked hawkes process for limit order book modeling,

    G. Chung, Y . Lee, and W. C. Kim, “Neural marked hawkes process for limit order book modeling,” in Advances in Knowledge Discovery and Data Mining - 28th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2024, Proceedings, Part III , 2024, pp. 197– 209

    work page 2024

  8. [8]

    Deep unsupervised learning using nonequilibrium thermodynamics,

    J. Sohl-Dickstein, E. A. Weiss, N. Maheswaranathan, and S. Ganguli, “Deep unsupervised learning using nonequilibrium thermodynamics,” in Proceedings of the 32nd International Conference on Machine Learning, ICML 2015, 2015, pp. 2256–2265

    work page 2015

Show all 39 references
  1. [9]

    Price dynamics in a markovian limit order market,

    R. Cont and A. de Larrard, “Price dynamics in a markovian limit order market,” SIAM J. Financial Math. , vol. 4, no. 1, pp. 1–25, 2013

    2013

  2. [10]

    Modelling trades-through in a limit order book using hawkes processes,

    I. M. Toke and F. Pomponio, “Modelling trades-through in a limit order book using hawkes processes,” Economics, vol. 6, no. 1, pp. 2012–22, 2012

    2012

  3. [11]

    A non-linear model of limit order book dynamics,

    N. D. Vvedenskaya, Y . M. Suhov, and V . Belitsky, “A non-linear model of limit order book dynamics,” in 2011 IEEE International Symposium on Information Theory Proceedings, ISIT 2011 , 2011, pp. 1260–1262

    2011

  4. [12]

    A dynamic model of the limit order book,

    I. Rocsu, “A dynamic model of the limit order book,” The Review of Financial Studies, vol. 22, no. 11, pp. 4601–4641, 2009

    2009

  5. [13]

    Price dynamics in limit order markets,

    C. A. Parlour, “Price dynamics in limit order markets,” The Review of Financial Studies, vol. 11, no. 4, pp. 789–816, 1998

    1998

  6. [14]

    High frequency trading strategies, market fragility and price spikes: an agent based model perspective,

    F. McGroarty, A. Booth, E. H. Gerding, and V . L. R. Chinthalapati, “High frequency trading strategies, market fragility and price spikes: an agent based model perspective,” Ann. Oper. Res., vol. 282, no. 1-2, pp. 217–244, 2019

    2019

  7. [15]

    Deeplob: Deep convolutional neural networks for limit order books,

    Z. Zhang, S. Zohren, and S. J. Roberts, “Deeplob: Deep convolutional neural networks for limit order books,” IEEE Trans. Signal Process. , vol. 67, no. 11, pp. 3001–3012, 2019

    2019

  8. [16]

    Towards realistic market simulations: a generative adversarial networks approach,

    A. Coletta, M. Prata, M. Conti, E. Mercanti, N. Bartolini, A. Moulin, S. Vyetrenko, and T. Balch, “Towards realistic market simulations: a generative adversarial networks approach,” in 2nd ACM International Conference on AI in Finance, 2021 , 2021, pp. 1–9

    2021

  9. [17]

    Generating real- istic stock market order streams,

    J. Li, X. Wang, Y . Lin, A. Sinha, and M. P. Wellman, “Generating real- istic stock market order streams,” in The Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2020 , 2020, pp. 727–734

    2020

  10. [18]

    The limit order book recreation model (LO- BRM): an extended analysis,

    Z. Shi and J. Cartlidge, “The limit order book recreation model (LO- BRM): an extended analysis,” in Machine Learning and Knowledge Discovery in Databases. Applied Data Science Track - European Con- ference, ECML PKDD 2021, Proceedings, Part IV , 2021, pp. 204–220

    2021

  11. [19]

    The LOB recreation model: Predicting the limit order book from TAQ history using an ordinary differential equation recurrent neural network,

    Z. Shi, Y . Chen, and J. Cartlidge, “The LOB recreation model: Predicting the limit order book from TAQ history using an ordinary differential equation recurrent neural network,” in Thirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021 , 2021, pp. 548–556

    2021

  12. [20]

    High- resolution image synthesis with latent diffusion models,

    R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High- resolution image synthesis with latent diffusion models,” in IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022 , 2022, pp. 10 674–10 685

    2022

  13. [21]

    Diffusion pixelation: A game diffusion model of rumor & anti-rumor inspired by image restoration,

    Y . Xiao, Z. Huang, Q. Li, X. Lu, and T. Li, “Diffusion pixelation: A game diffusion model of rumor & anti-rumor inspired by image restoration,” IEEE Trans. Knowl. Data Eng. , vol. 35, no. 5, pp. 4682– 4694, 2023

    2023

  14. [22]

    Non-autoregressive conditional diffusion models for time series prediction,

    L. Shen and J. T. Kwok, “Non-autoregressive conditional diffusion models for time series prediction,” in International Conference on Machine Learning, ICML 2023 , 2023, pp. 31 016–31 029

    2023

  15. [23]

    Diffusion variational autoencoder for tackling stochasticity in multi-step regression stock price prediction,

    K. J. L. Koa, Y . Ma, R. Ng, and T. Chua, “Diffusion variational autoencoder for tackling stochasticity in multi-step regression stock price prediction,” in Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, CIKM 2023 , 2023, pp. 1087–1096

    2023

  16. [24]

    Pristi: A conditional diffusion framework for spatiotemporal imputation,

    M. Liu, H. Huang, H. Feng, L. Sun, B. Du, and Y . Fu, “Pristi: A conditional diffusion framework for spatiotemporal imputation,” in 39th IEEE International Conference on Data Engineering, ICDE 2023, 2023, pp. 1927–1939

    2023

  17. [25]

    Imdiffusion: Imputed diffusion models for mul- tivariate time series anomaly detection,

    Y . Chen, C. Zhang, M. Ma, Y . Liu, R. Ding, B. Li, S. He, S. Rajmohan, Q. Lin, and D. Zhang, “Imdiffusion: Imputed diffusion models for mul- tivariate time series anomaly detection,” Proc. VLDB Endow. , vol. 17, no. 3, pp. 359–372, 2023

    2023

  18. [26]

    Scope- free global multi-condition-aware industrial missing data imputation framework via diffusion transformer,

    D. Liu, Y . Wang, C. Liu, X. Yuan, K. Wang, and C. Yang, “Scope- free global multi-condition-aware industrial missing data imputation framework via diffusion transformer,” IEEE Trans. Knowl. Data Eng. , vol. 36, no. 11, pp. 6977–6988, 2024

    2024

  19. [27]

    A survey on generative diffusion models,

    H. Cao, C. Tan, Z. Gao, Y . Xu, G. Chen, P. Heng, and S. Z. Li, “A survey on generative diffusion models,” IEEE Trans. Knowl. Data Eng., vol. 36, no. 7, pp. 2814–2830, 2024

    2024

  20. [28]

    Controllable tabular data synthesis using diffusion models,

    T. Liu, J. Fan, N. Tang, G. Li, and X. Du, “Controllable tabular data synthesis using diffusion models,” Proc. ACM Manag. Data , vol. 2, no. 1, pp. 28:1–28:29, 2024

    2024

  21. [29]

    DOSE: diffusion dropout with adaptive prior for speech enhancement,

    W. Tai, Y . Lei, F. Zhou, G. Trajcevski, and T. Zhong, “DOSE: diffusion dropout with adaptive prior for speech enhancement,” in Advances in Neural Information Processing Systems, NeurIPS 2023 , 2023

    2023

  22. [30]

    Diffsformer: A diffusion transformer on stock factor augmentation,

    Y . Gao, H. Chen, X. Wang, Z. Wang, X. Wang, J. Gao, and B. Ding, “Diffsformer: A diffusion transformer on stock factor augmentation,” CoRR, vol. abs/2402.06656, 2024

    2024 arXiv

  23. [31]

    Friedman, The double auction market: institutions, theories, and evidence

    D. Friedman, The double auction market: institutions, theories, and evidence. Routledge, 2018

    2018

  24. [32]

    Statistical properties of stock order books: empirical results and models,

    J.-P. Bouchaud, M. M ´ezard, and M. Potters, “Statistical properties of stock order books: empirical results and models,” Quantitative finance, vol. 2, no. 4, p. 251, 2002

    2002

  25. [33]

    Denoising diffusion probabilistic models,

    J. Ho, A. Jain, and P. Abbeel, “Denoising diffusion probabilistic models,” in Advances in Neural Information Processing Systems, NeurIPS 2020 , 2020, pp. 6840–6851

    2020

  26. [34]

    Attention is all you need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, “Attention is all you need,” in Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017 , 2017, pp. 5998–6008

    2017

  27. [35]

    gddim: Generalized denoising diffusion implicit models,

    Q. Zhang, M. Tao, and Y . Chen, “gddim: Generalized denoising diffusion implicit models,” in The Eleventh International Conference on Learning Representations, ICLR 2023 , 2023

    2023

  28. [36]

    Denoising diffusion implicit models,

    J. Song, C. Meng, and S. Ermon, “Denoising diffusion implicit models,” in 9th International Conference on Learning Representations, ICLR 2021, 2021

    2021

  29. [37]

    State-dependent hawkes pro- cesses and their application to limit order book modelling,

    M. Morariu-Patrichi and M. S. Pakkanen, “State-dependent hawkes pro- cesses and their application to limit order book modelling,” Quantitative Finance, vol. 22, no. 3, pp. 563–583, 2022

    2022

  30. [38]

    Long short-term memory,

    S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Comput., vol. 9, no. 8, pp. 1735–1780, 1997

    1997

  31. [39]

    Neighbor profile: Bagging nearest neigh- bors for unsupervised time series mining,

    Y . He, X. Chu, and Y . Wang, “Neighbor profile: Bagging nearest neigh- bors for unsupervised time series mining,” in 36th IEEE International Conference on Data Engineering, ICDE 2020 , 2020, pp. 373–384

    2020

Pith tools

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