Pith. sign in

REVIEW 3 major objections 6 minor 5 cited by

Multimodal Conditioned Diffusive Time Series Forecasting

T0 review · 3 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read MCD-TSF conditions a denoising diffusion network on timestamps and text, and the paper reports that this joint conditioning achieves the lowest average MSE and MAE across a benchmark of eight domains.

desk verdict A sensible integration of timestamps and text into diffusion forecasting that probably improves average performance, but the SOTA claim outruns the reported statistics. read the letter →

arxiv 2504.19669 v1 pith:VX5MKC2A submitted 2025-04-28 cs.CL

Chen Su , Yuanhe Tian , Yan Song This is my paper · ORCID
classification cs.CL
keywords diffusionmodelstimeseriesforecastingmultimodalconditioningclassifier-freeguidancetimestampencodingtext-conditionedprobabilisticTransformerfusion
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 claims that time-series forecasting improves when a diffusion model is conditioned not only on the numeric history but also on timestamps and free-text descriptions, through a model it calls MCD-TSF. Its denoising network first fuses timestamp features into the series with attention, then fuses text through cross-attention, and uses classifier-free guidance to control how strongly text steers each step. On benchmarks spanning agriculture, climate, economy, energy, environment, health, social good, and traffic, the paper reports lower MSE and MAE than diffusion-only, text-only, timestamp-only, and existing multimodal forecasters on most domains. If the claim holds, it would mean that metadata already stored alongside many series is enough to sharpen probabilistic forecasts without converting the series into text for a large language model.

What carries the argument

The carrying mechanism is the denoising network $f_\theta$ with a multimodal fusion stack. Each layer runs timestamp-assisted attention (TAA), which concatenates timestamp embeddings with the series representation and applies multi-head self-attention so structural time semantics propagate between data points, followed by text-time fusion (TTF), cross-attention in which series features are queries and text embeddings are keys and values. Classifier-free guidance provides the control law: at each denoising step the model predicts with text and without text, and the actual estimate is the text-free prediction plus $w$ times their difference. This lets $w$ tune how much textual semantics shape the forecast while keeping the timestamp-only path intact.

What would settle it

Re-run MCD-TSF and its closest competitors, for instance PatchTST on climate and GLAFF on social good, many times from different seeds and compare the per-seed MSE distributions; if the intervals overlap where Table 3 shows near ties, the claimed advantage is not distinguishable from noise.

Watch

Extended reading notes

Core claim

The paper's central claim is that a diffusion forecaster can be improved by conditioning the reverse denoising process on two extra modalities: timestamps and natural-language text. MCD-TSF encodes the numeric history, timestamp features such as day-of-week and month-of-year, and text from a frozen language model; its fusion layers first inject timestamp structure through timestamp-assisted attention and then absorb text through text-series cross-attention. At inference, classifier-free guidance linearly mixes the text-conditioned prediction with a text-free prediction via a strength weight w, so the model can damp irrelevant text. On real-world benchmarks spanning eight domains, the paper reports the lowest average MSE (0.638) and MAE (0.484) among all compared models, including diffusion, transformer, MLP, state-space, LLM, text-integrated, and timestamp-integrated baselines.

Load-bearing premise

The paper's state-of-the-art claim rests on the assumption that averaging three random-seed runs without reporting variance is enough to certify small performance gaps, since several best-versus-runner-up differences in Table 3 are close to run-to-run noise.

Editorial extensions

If this is right

  • Adding either timestamps or text to plain diffusion improves average MSE (DIFF+TAA at 0.794 and DIFF+TTF at 0.796 versus DIFF at 0.904), and using both with the proposed fusion lowers it further to 0.638.
  • The classifier-free weight $w$ gives a single inference-time dial for text influence; across most domains the error curve is U-shaped, so the model can be tuned to avoid both ignoring text and over-trusting it.
  • The model degrades gracefully when text is absent or uninformative: setting $w$ to 0 recovers the timestamp-only behavior, matching the paper's stated robustness goal for missing-text inputs.
  • Higher timestamp weight $\lambda$ improves performance in the tested domains, supporting the paper's claim that temporal structure carries forecasting signal beyond the raw series values.

Reading between the lines

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

  • Not tested in the paper: because the text encoder is a frozen BERT-base, a natural extension is to check whether a domain-adapted or larger language model changes the optimal $w$ and the size of the text-conditioning gain.
  • Not tested in the paper: the same fusion pattern could accept other metadata such as weather alerts, calendar events, or geolocation in place of text, with the classifier-free weight controlling that modality's influence.
  • The sparse-text environment domain (4.2% coverage) shows little text benefit, which suggests that in deployment the model's gains will concentrate where auxiliary descriptions are dense; retrieval of similar past reports might recover gains in sparse-text domains.
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. This paper proposes MCD-TSF, a diffusion-based time series forecasting model that conditions the reverse denoising process on the historical series, timestamps, and text. The architecture combines a multimodal encoder, timestamp-assisted attention (TAA), text-time series fusion (TTF), and classifier-free guidance to control the influence of text during inference. The paper evaluates on the Time-MMD benchmark across eight domains, reporting MSE/MAE averaged over three seeds, and claims state-of-the-art performance. Ablations study the timestamp weight, text guidance strength, and unconditional training probability.

Significance. The modeling direction is timely: combining probabilistic diffusion forecasting with timestamp and text conditioning is a sensible and potentially useful extension of existing unimodal diffusion forecasters. If the empirical advantages hold, the paper demonstrates a practical recipe for multimodal probabilistic forecasting that avoids full-scale LLM inference by using a frozen BERT encoder. Strengths include the open-source code link, a standard diffusion formulation, validation-based hyperparameter selection, and ablations that isolate the contribution of each modality. The central SOTA claim, however, currently rests on averaged point estimates with no variance information, and several per-domain comparisons in Table 3 do not favor MCD-TSF, including a large reversal on Climate. The claim is therefore conditional on additional statistical evidence and on clarification of the fusion and evaluation equations.

major comments (3)
  1. [Section 4.3, Table 3] The abstract's state-of-the-art claim is not yet supported by the reported statistics. Section 4.3 states only that "all experiments are repeated three times with different random seeds, and final results are the average of the three runs," with no standard deviation, per-seed table, or significance test. In Table 3, several comparisons show that MCD-TSF is not the best model per domain: on Climate, DLinear reports MSE 0.950 versus MCD-TSF's 1.583; on Environment, PatchTST reports 0.254 versus 0.275; and on Health, MM-TSF reports 1.368 versus 1.496. The average advantage (0.638 Avg MSE versus PatchTST's 0.685) cannot be distinguished from seed noise without variance information for the near-tie cases, and the Climate reversal is too large to be explained by the missing variance alone. The same issue affects the hyperparameter comparisons in Sections 5.2 and 5.3. Please report per-seed results (or at least standard deviations), perform a paired significance test or equivalent, and qualify the SOTA claim so that it is not read as per-domain superiority.
  2. [Equation (16)] The fusion equation is ambiguous and appears to contain an unexplained term. The text reads "bYk-1 = X W ⊙ (bYs ⊕ bYu)", with W = MLP(bXu − X) from Eq. (15). The leading X is unexplained, and the shapes are not defined: W is derived from bXu − X, whose length is not specified, while the Hadamard product is applied to the concatenation bYs ⊕ bYu. As written, the operation cannot be implemented unambiguously. Please clarify whether a summation or a different fusion operation is intended, and state the exact dimensions of W, bYs, bYu, and the output.
  3. [Appendix D, Eq. (19)] The evaluation metric is mis-defined. Equation (19) prints MSE as sqrt(mean(|bY − Y*|)) (or a square-root mean absolute error), which is neither the mean squared error used in the main text nor the metric used by the cited baselines. This is load-bearing because the central comparison rests on this metric. Please correct the formula to MSE = mean((bY − Y*)^2) (or explicitly state and consistently apply an RMSE definition) and verify that all reported numbers use the corrected definition.
minor comments (6)
  1. [Section 4.3 vs. Appendix A and Table 4] The timestamp feature dimensions are stated inconsistently. Section 4.3 says monthly, weekly, and daily frequencies use dimensions three, two, and one, respectively, while Appendix A and Table 4 assign one, two, and three dimensions to monthly, weekly, and daily data. One of these statements is inverted; please correct the inconsistency and state which configuration was used.
  2. [Equation (13)] The pooling operation uses 1/sqrt(L) as the scaling factor but is described as mean pooling. If the scaling is intentional, please justify it; otherwise it should be 1/L for a true mean.
  3. [Figure 3 caption] The caption says darker colors correspond to higher attention weights, while the main text says brighter colors stand for higher attention weights. Please align the caption with the text and the colormap.
  4. [Section 4.3 and Table 1] For domains with low text coverage (e.g., Agriculture 13.3% and Environment 4.2%), the paper does not specify how missing reports are represented in the concatenated text and how this interacts with the unconditional input ∅ in Eq. (12). Please clarify whether missing text becomes an empty string and whether such samples are treated as unconditional during training.
  5. [Tables 2 and 3] The caption of Table 2 states that the first and second best results are marked by boldface and underlines, but Table 3, which also appears to mark best results, contains no such markers. Please apply the markers consistently or remove the statement.
  6. [Appendix F.2] The cross-reference to Tables 6 and 7 appears as "Table ??" in the text. Please fix the reference.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: MCD-TSF is an empirical architecture trained on external benchmarks with validation-based hyperparameter selection, and the paper's few self-citations are peripheral.

full rationale

The paper contains no derivation chain in which an output is equivalent to its input by construction. The model is defined in Section 3 from inputs X, U, and E, trained with the conditional MSE loss in Eq. (11), and evaluated on held-out test partitions of the external Time-MMD benchmark. Key hyperparameters (lambda, w, p_uncond) are selected on the validation set (Section 4.3), and the learned fusion weight W in Eqs. (15)-(16) is part of the trained model, not a constant fitted to the reported test numbers. The bespoke ablations DIFF+TAA-T and DIFF+TTF-T serve as baselines, not as inputs to a derivation. The paper's self-citations ([45], [51], [52]) support only peripheral claims about pretrained encoders and diffusion models and are not load-bearing. The strongest concern, that three-seed averages without variance leave near-tie SOTA margins uncertain (Table 3), is a statistical robustness and correctness issue, not circularity. The unresolved cross-reference 'Table ??' in Appendix F.2 is a presentation defect, not a circular step.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The model relies on standard diffusion math, a frozen BERT encoder, and several hand-tuned hyperparameters. No new physical entities are introduced. The central empirical claim depends on dataset assumptions about timestamp utility and text relevance, plus unvalidated architectural choices in the fusion layer.

free parameters (4)
  • textual guidance strength w = 0.8 (default, tuned over [0.1, 2.5])
    Controls the text contribution in the classifier-free guidance combination in Eq. (12); selected based on validation set performance.
  • timestamp weight lambda = 1.0 (default, tuned over [0.2, 1.0])
    Weights the timestamp embeddings in the TAA module; higher values improved performance across tested domains.
  • unconditional training probability puncond = 0.1 (default, tuned over [0.05, 0.7])
    Probability of masking text during training to enable classifier-free guidance; tuned on the validation set.
  • model hyperparameters (layers, heads, hidden dimension) = 6 layers, 8 heads, 64 hidden dimension
    Architectural choices validated empirically, not derived from first principles.
assumptions (6)
  • standard math DDPM forward and reverse diffusion equations
    Eqs. (1)-(8) follow Ho et al. (2020) and Tashiro et al. (2021).
  • standard math Classifier-free guidance formula
    Eq. (12) follows Ho and Salimans (2022).
  • domain assumption Timestamps encode useful temporal structure
    The TAA module assumes calendar features such as day-of-week and month help forecasting; this is supported only by ablations, not by a formal argument.
  • domain assumption Text descriptions provide useful conditioning information
    The TTF module assumes BERT embeddings of concatenated reports improve forecasts; in low-coverage domains such as Environment (4.2% text coverage) this is questionable.
  • ad hoc to paper Concatenating 36 intervals of text into one document is a sufficient summary
    Appendix B defines the template and aggregation rule; no theoretical or empirical justification is given for the 36-interval window.
  • ad hoc to paper Learned fusion weight W = MLP(bXu - X) is a valid gating mechanism
    Eqs. (15)-(16) introduce an ad hoc weighting of two prediction pathways based on the timestamp-only reconstruction error; no first-principles derivation is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multimodal Conditioned Diffusive Time Series Forecasting." pith.science (2026). https://pith.science/paper/VX5MKC2A

@misc{pith2026250419669,
  author       = {Pith},
  title        = {Pith review of: Multimodal Conditioned Diffusive Time Series Forecasting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VX5MKC2A}},
  note         = {Machine review of arXiv:2504.19669}
}
read the original abstract

Diffusion models achieve remarkable success in processing images and text, and have been extended to special domains such as time series forecasting (TSF). Existing diffusion-based approaches for TSF primarily focus on modeling single-modality numerical sequences, overlooking the rich multimodal information in time series data. To effectively leverage such information for prediction, we propose a multimodal conditioned diffusion model for TSF, namely, MCD-TSF, to jointly utilize timestamps and texts as extra guidance for time series modeling, especially for forecasting. Specifically, Timestamps are combined with time series to establish temporal and semantic correlations among different data points when aggregating information along the temporal dimension. Texts serve as supplementary descriptions of time series' history, and adaptively aligned with data points as well as dynamically controlled in a classifier-free manner. Extensive experiments on real-world benchmark datasets across eight domains demonstrate that the proposed MCD-TSF model achieves state-of-the-art performance.

Figures

Figures reproduced from arXiv: 2504.19669 by the authors.

Figure 1
Figure 1. The overall architecture of the proposed MCD-TSF. Our approach consists of four com [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Results of our approach on different domains when configured with various values of the [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Visual of the attention of our approach across data points of two cases with different [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Figure (a) illustrates how the model’s MSE varies with different guidance strengths across [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Case study with four examples from energy domains. In all cases, the historical time series and the gold standard future time series are marked by blue and dashed grey lines, respectively. The predictions from PatchTST, CSDI, and our approach (i.e., MCD-TSF) are repres…
Figure 6
Figure 6. Figure 6: Results of our approach on different domains when configured with various values of the [PITH_FULL_IMAGE:figures/full_fig_p021_6.png]
Figure 7
Figure 7. Figure 7: The model performance with changes in the unconditional training probability. [PITH_FULL_IMAGE:figures/full_fig_p022_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

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

  1. Differencing the Diffusion Trajectory toward Uncertain Components for Time Series Forecasting

    cs.AI 2026-06 conditional novelty 6.0 of 10

    DiffDiff rewires diffusion forecasting so corruption gradually emphasizes second-order differences, concentrating generation on history-uncertain parts and improving forecasts on seven benchmarks.

  2. Rethinking Multimodal Fusion for Time Series: Text Modalities Need Constrained Fusion

    cs.LG 2026-03 unverdicted novelty 6.0 of 10

    Uncontrolled text–time-series fusion underperforms unimodal baselines; constrained fusion and a low-rank Controlled Fusion Adapter recover gains without changing the TS backbone.

  3. Text Reinforcement for Multimodal Time Series Forecasting

    cs.CL 2025-08 conditional novelty 6.0 of 10

    Reinforcement learning trains an LLM to generate improved text from time series, improving multimodal forecasting on Time-MMD.

  4. Fusing Large Language Models with Temporal Transformers for Time Series Forecasting

    cs.CL 2025-07 conditional novelty 5.0 of 10

    A gated fusion of GPT-2 semantic features and a PatchTST-style Transformer encoder improves average MSE/MAE slightly on ETT, Weather, and ILI, while losing to PatchTST on four of the six datasets.

  5. Diffusion Models for Time Series Forecasting: A Survey

    stat.ML 2025-07 conditional novelty 4.0 of 10

    A survey classifies diffusion-based time series forecasting models into a two-axis taxonomy by conditioning source and integration method.

Reference graph

Works this paper leans on

72 extracted references · 49 canonical work pages · cited by 5 Pith papers

  1. [1]

    Improving Time Series Forecasting Using LSTM and Attention Models

    Hossein Abbasimehr and Reza Paki. Improving Time Series Forecasting Using LSTM and Attention Models. Journal of Ambient Intelligence and Humanized Computing, 13(1):673–691, 2022

    work page 2022

  2. [2]

    GPT-4 Technical Report

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. GPT-4 Technical Report. arXiv preprint arXiv:2303.08774, 2023

    arXiv 2023

  3. [3]

    TimeMachine: A Time Series Is Worth 4 Mambas for Long-Term Forecasting

    Md Atik Ahamed and Qiang Cheng. TimeMachine: A Time Series Is Worth 4 Mambas for Long-Term Forecasting. In 27th European Conference on Artificial Intelligence, volume 392, page 1688, 2024

    work page 2024

  4. [4]

    Stock Price Prediction Using the ARIMA Model

    Adebiyi A Ariyo, Adewumi O Adewumi, and Charles K Ayo. Stock Price Prediction Using the ARIMA Model. In 2014 UKSim-AMSS 16th International Conference on Computer Modelling and Simulation, pages 106–112. IEEE, 2014

    work page 2014

  5. [5]

    Conditional Time Series Forecasting With Convolutional Neural Networks

    Anastasia Borovykh, Sander Bohte, and Cornelis W Oosterlee. Conditional Time Series Forecasting With Convolutional Neural Networks. arXiv preprint arXiv:1703.04691, 2017

    arXiv 2017

  6. [6]

    Hybrid Variational Autoencoder for Time Series Forecasting

    Borui Cai, Shuiqiao Yang, Longxiang Gao, and Yong Xiang. Hybrid Variational Autoencoder for Time Series Forecasting. Knowledge-Based Systems, 281:111079, 2023

    work page 2023

  7. [7]

    Financial Time Series Forecasting Model Based on CEEMDAN and LSTM

    Jian Cao, Zhi Li, and Jian Li. Financial Time Series Forecasting Model Based on CEEMDAN and LSTM. Physica A: Statistical Mechanics and Its Applications, 519:127–139, 2019

    work page 2019

  8. [8]

    Advanced LSTM-Based Time Series Forecasting for Enhanced Energy Consumption Management in Electric Power Systems

    VS Chandrika, NMG Kumar, Vinjamuri Venkata Kamesh, A Shobanadevi, V Maheswari, K Sekar, T Logeswaran, and A Rajaram. Advanced LSTM-Based Time Series Forecasting for Enhanced Energy Consumption Management in Electric Power Systems. International Journal of Renewable Energy Research (IJRER), 14(1):127–139, 2024

    work page 2024

Show all 72 references
  1. [9]

    Financial Time Series Forecasting With Multi-Modality Graph Neural Network

    Dawei Cheng, Fangzhou Yang, Sheng Xiang, and Jin Liu. Financial Time Series Forecasting With Multi-Modality Graph Neural Network. Pattern Recognition, 121:108218, 2022. 13

    2022

  2. [10]

    Understanding Different Design Choices in Training Large Time Series Models

    Yu-Neng Chuang, Songchen Li, Jiayi Yuan, Guanchu Wang, Kwei-Herng Lai, Leisheng Yu, Sirui Ding, Chia-Yuan Chang, Qiaoyu Tan, Daochen Zha, et al. Understanding Different Design Choices in Training Large Time Series Models. arXiv preprint arXiv:2406.14045, 2024

    2024

  3. [11]

    ARIMA Models to Predict Next-Day Electricity Prices

    Javier Contreras, Rosario Espinola, Francisco J Nogales, and Antonio J Conejo. ARIMA Models to Predict Next-Day Electricity Prices. IEEE Transactions on Power Systems, 18(3):1014–1020, 2003

    2003

  4. [12]

    Bert: Pre-training of Deep Bidirectional Transformers for Language Understanding

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human langu...

    2019

  5. [13]

    Preformer: Predictive Transformer With Multi-Scale Segment-Wise Correlations for Long-Term Time Series Forecasting

    Dazhao Du, Bing Su, and Zhewei Wei. Preformer: Predictive Transformer With Multi-Scale Segment-Wise Correlations for Long-Term Time Series Forecasting. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE, 2023

    2023

  6. [14]

    Diffusion-Driven Incomplete Multimodal Learning for Air Quality Prediction

    Jinxiao Fan, Mengshi Qi, Liang Liu, and Huadong Ma. Diffusion-Driven Incomplete Multimodal Learning for Air Quality Prediction. ACM Transactions on Internet of Things, 6(1):1–24, 2025

    2025

  7. [15]

    MG-TSD: Multi-granularity Time Series Diffusion Models with Guided Learning Process

    Xinyao Fan, Yueying Wu, Chang Xu, Yuhao Huang, Weiqing Liu, and Jiang Bian. MG-TSD: Multi-granularity Time Series Diffusion Models with Guided Learning Process. arXiv preprint arXiv:2403.05751, 2024

    2024 arXiv

  8. [16]

    Multi-Scale Attention Flow for Probabilistic Time Series Forecasting

    Shibo Feng, Chunyan Miao, Ke Xu, Jiaxiang Wu, Pengcheng Wu, Yang Zhang, and Peilin Zhao. Multi-Scale Attention Flow for Probabilistic Time Series Forecasting. IEEE Transactions on Knowledge and Data Engineering, 36(5):2056–2068, 2023

    2023

  9. [17]

    The LLaMA 3 Herd of Models

    Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, et al. The LLaMA 3 Herd of Models. arXiv preprint arXiv:2407.21783, 2024

    2024 arXiv

  10. [19]

    Large Language Models Are Zero-Shot Time Series Forecasters

    Nate Gruver, Marc Finzi, Shikai Qiu, and Andrew G Wilson. Large Language Models Are Zero-Shot Time Series Forecasters. Advances in Neural Information Processing Systems , 36:19622–19635, 2023

    2023

  11. [20]

    Denoising Diffusion Probabilistic Models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising Diffusion Probabilistic Models. In Neural Information Processing Systems, volume 33, pages 6840–6851, 2020

    2020

  12. [21]

    Classifier-free diffusion guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598, 2022

    2022 arXiv

  13. [22]

    Times Series Forecasting for Urban Building Energy Consumption Based on Graph Convolutional Network

    Yuqing Hu, Xiaoyuan Cheng, Suhang Wang, Jianli Chen, Tianxiang Zhao, and Enyan Dai. Times Series Forecasting for Urban Building Energy Consumption Based on Graph Convolutional Network. Applied Energy, 307:118231, 2022

    2022

  14. [23]

    WITran: Water-Wave Information Transmission and Recurrent Acceleration Network for Long-Range Time Series Forecasting

    Yuxin Jia, Youfang Lin, Xinyan Hao, Yan Lin, Shengnan Guo, and Huaiyu Wan. WITran: Water-Wave Information Transmission and Recurrent Acceleration Network for Long-Range Time Series Forecasting. Advances in Neural Information Processing Systems, 36:12389–12356, 2023

    2023

  15. [24]

    Time-LLM: Time Series Forecasting by Reprogramming Large Language Models

    Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y Zhang, Xiaoming Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, et al. Time-LLM: Time Series Forecasting by Reprogramming Large Language Models. arXiv preprint arXiv:2310.01728, 2023

    arXiv

  16. [25]

    Reformer: The Efficient Transformer

    Nikita Kitaev, Łukasz Kaiser, and Anselm Levskaya. Reformer: The Efficient Transformer. arXiv preprint arXiv:2001.04451, 2020

    2001 arXiv

  17. [26]

    LightCTS: A Lightweight Framework for Correlated Time Series Forecasting

    Zhichen Lai, Dalin Zhang, Huan Li, Christian S Jensen, Hua Lu, and Yan Zhao. LightCTS: A Lightweight Framework for Correlated Time Series Forecasting. Proceedings of the ACM on Management of Data, 1(2):1–26, 2023

    2023

  18. [27]

    Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglement

    Yan Li, Xinjiang Lu, Yaqing Wang, and Dejing Dou. Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglement. In Neural Information Processing Systems, volume 35, pages 23009–23022, 2022. 14

    2022

  19. [28]

    Time-MMD: Multi-Domain Multimodal Dataset for Time Series Analysis

    Haoxin Liu, Shangqing Xu, Zhiyuan Zhao, Lingkai Kong, Harshavardhan Prabhakar Kamarthi, Aditya Sasanur, Megha Sharma, Jiaming Cui, Qingsong Wen, Chao Zhang, et al. Time-MMD: Multi-Domain Multimodal Dataset for Time Series Analysis. Advances in Neural Information Processing Sys...

    2024

  20. [29]

    Retrieval-Augmented Diffusion Models for Time Series Forecasting

    Jingwei Liu, Ling Yang, Hongyan Li, and Shenda Hong. Retrieval-Augmented Diffusion Models for Time Series Forecasting. Advances in Neural Information Processing Systems , 37:2766–2786, 2024

    2024

  21. [30]

    Pyraformer: Low-Complexity Pyramidal Attention for Long-Range Time Series Modeling and Forecasting

    Shizhan Liu, Hang Yu, Cong Liao, Jianguo Li, Weiyao Lin, Alex X Liu, and Schahram Dustdar. Pyraformer: Low-Complexity Pyramidal Attention for Long-Range Time Series Modeling and Forecasting. In International Conference on Learning Representations, 2022

    2022

  22. [31]

    Unitime: A Language-Empowered Unified Model for Cross-Domain Time Series Forecasting

    Xu Liu, Junfeng Hu, Yuan Li, Shizhe Diao, Yuxuan Liang, Bryan Hooi, and Roger Zimmermann. Unitime: A Language-Empowered Unified Model for Cross-Domain Time Series Forecasting. In Proceedings of the ACM Web Conference 2024, pages 4095–4106, 2024

    2024

  23. [32]

    AutoTimes: Autoregressive Time Series Forecasters via Large Language Models

    Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, and Mingsheng Long. AutoTimes: Autoregressive Time Series Forecasters via Large Language Models. Advances in Neural Information Processing Systems, 37:122154–122184, 2024

    2024

  24. [33]

    Short-Term Traffic Flow Prediction for Urban Road Sections Based on Time Series Analysis and LSTM-BiLSTM Method

    Changxi Ma, Guowen Dai, and Jibiao Zhou. Short-Term Traffic Flow Prediction for Urban Road Sections Based on Time Series Analysis and LSTM-BiLSTM Method. IEEE Transactions on Intelligent Transportation Systems, 23(6):5615–5624, 2021

    2021

  25. [34]

    Short-Term Traffic Flow Prediction for Urban Road Sections Based on Time Series Analysis and LSTM_BILSTM Method

    Changxi Ma, Guowen Dai, and Jibiao Zhou. Short-Term Traffic Flow Prediction for Urban Road Sections Based on Time Series Analysis and LSTM_BILSTM Method. IEEE Transactions on Intelligent Transportation Systems, 23(6):5615–5624, 2021

    2021

  26. [35]

    A Time Series Is Worth 64 Words: Long-Term Forecasting with Transformers

    Yuqi Nie, Nam H Nguyen, Phanwadee Sinthong, and Jayant Kalagnanam. A Time Series Is Worth 64 Words: Long-Term Forecasting with Transformers. arXiv preprint arXiv:2211.14730, 2022

    2022 arXiv

  27. [36]

    S2IP-LLM: Semantic Space Informed Prompt Learning With LLM for Time Series Forecasting

    Zijie Pan, Yushan Jiang, Sahil Garg, Anderson Schneider, Yuriy Nevmyvaka, and Dongjin Song. S2IP-LLM: Semantic Space Informed Prompt Learning With LLM for Time Series Forecasting. In Forty-First International Conference on Machine Learning, 2024

    2024

  28. [37]

    Ensemble Learning for Electricity Con- sumption Forecasting in Office Buildings

    Tiago Pinto, Isabel Praça, Zita Vale, and Jose Silva. Ensemble Learning for Electricity Con- sumption Forecasting in Office Buildings. Neurocomputing, 423:747–755, 2021

    2021

  29. [38]

    Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting

    Kashif Rasul, Calvin Seward, Ingmar Schuster, and Roland V ollgraf. Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting. In International Conference on Machine Learning, pages 8857–8868. PMLR, 2021

    2021

  30. [39]

    A Multi- Head Attention-Based Transformer Model for Traffic Flow Forecasting With a Comparative Analysis to Recurrent Neural Networks

    Selim Reza, Marta Campos Ferreira, José JM Machado, and João Manuel RS Tavares. A Multi- Head Attention-Based Transformer Model for Traffic Flow Forecasting With a Comparative Analysis to Recurrent Neural Networks. Expert Systems with Applications, 202:117275, 2022

    2022

  31. [40]

    Vision-Enhanced Time Series Forecasting via Latent Diffusion Models

    Weilin Ruan, Siru Zhong, Haomin Wen, and Yuxuan Liang. Vision-Enhanced Time Series Forecasting via Latent Diffusion Models. arXiv preprint arXiv:2502.14887, 2025

    2025 arXiv

  32. [41]

    Time Series Forecasting of Petroleum Production Using Deep LSTM Recurrent Networks

    Alaa Sagheer and Mostafa Kotb. Time Series Forecasting of Petroleum Production Using Deep LSTM Recurrent Networks. Neurocomputing, 323:203–213, 2019

    2019

  33. [42]

    FDNet: Focal Decomposed Network for Efficient, Robust and Practical Time Series Forecasting.Knowledge-Based Systems, 275:110666, 2023

    Li Shen, Yuning Wei, Yangzhu Wang, and Huaxin Qiu. FDNet: Focal Decomposed Network for Efficient, Robust and Practical Time Series Forecasting.Knowledge-Based Systems, 275:110666, 2023

    2023

  34. [43]

    Multi-Resolution Diffusion Models for Time Series Forecasting

    Lifeng Shen, Weiyu Chen, and James Kwok. Multi-Resolution Diffusion Models for Time Series Forecasting. In International Conference on Learning Representations, 2024

    2024

  35. [44]

    Denoising Diffusion Implicit Models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising Diffusion Implicit Models. arXiv preprint arXiv:2010.02502, 2020

    2010 arXiv

  36. [45]

    Zen 2.0: Continue Training and Adaption for N-gram Enhanced Text Encoders

    Yan Song, Tong Zhang, Yonggang Wang, and Kai-Fu Lee. Zen 2.0: Continue Training and Adaption for N-gram Enhanced Text Encoders. arXiv preprint arXiv:2105.01279, 2021

    2021 arXiv

  37. [47]

    TEST: Text Prototype Aligned Embedding to Activate LLM’s Ability for Time Series

    Chenxi Sun, Hongyan Li, Yaliang Li, and Shenda Hong. TEST: Text Prototype Aligned Embedding to Activate LLM’s Ability for Time Series. arXiv preprint arXiv:2308.08241, 2023. 15

    2023 arXiv

  38. [48]

    Hierarchical Classification Auxiliary Network for Time Series Forecasting.arXiv preprint arXiv:2405.18975, 2024

    Yanru Sun, Zongxia Xie, Dongyue Chen, Emadeldeen Eldele, and Qinghua Hu. Hierarchical Classification Auxiliary Network for Time Series Forecasting.arXiv preprint arXiv:2405.18975, 2024

    2024 arXiv

  39. [49]

    Time Series Forecasting With LLMs: Understanding and Enhancing Model Capabilities

    Hua Tang, Chong Zhang, Mingyu Jin, Qinkai Yu, Zhenting Wang, Xiaobo Jin, Yongfeng Zhang, and Mengnan Du. Time Series Forecasting With LLMs: Understanding and Enhancing Model Capabilities. ACM SIGKDD Explorations Newsletter, 26(2):109–118, 2025

    2025

  40. [50]

    CSDI: Conditional Score-Based Diffusion Models for Probabilistic Time Series Imputation

    Yusuke Tashiro, Jiaming Song, Yang Song, and Stefano Ermon. CSDI: Conditional Score-Based Diffusion Models for Probabilistic Time Series Imputation. In Neural Information Processing Systems, volume 34, pages 24804–24816, 2021

    2021

  41. [51]

    Chimed-GPT: A Chinese Medical Large Language Model With Full Training Regime and Better Alignment to Human Preferences

    Yuanhe Tian, Ruyi Gan, Yan Song, Jiaxing Zhang, and Yongdong Zhang. Chimed-GPT: A Chinese Medical Large Language Model With Full Training Regime and Better Alignment to Human Preferences. arXiv preprint arXiv:2311.06025, 2023

    2023 arXiv

  42. [52]

    Diffusion Networks with Task-Specific Noise Control for Radiology Report Generation

    Yuanhe Tian, Fei Xia, and Yan Song. Diffusion Networks with Task-Specific Noise Control for Radiology Report Generation. In Proceedings of the 32nd ACM International Conference on Multimedia, pages 1771–1780, 2024

    2024

  43. [53]

    Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspective

    Chengsen Wang, Qi Qi, Jingyu Wang, Haifeng Sun, Zirui Zhuang, Jinming Wu, and Jianxin Liao. Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspective. arXiv preprint arXiv:2409.18696, 2024

    2024 arXiv

  44. [54]

    ChatTime: A Unified Multimodal Time Series Foundation Model Bridging Numerical and Textual Data

    Chengsen Wang, Qi Qi, Jingyu Wang, Haifeng Sun, Zirui Zhuang, Jinming Wu, Lei Zhang, and Jianxin Liao. ChatTime: A Unified Multimodal Time Series Foundation Model Bridging Numerical and Textual Data. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, ...

    2025

  45. [55]

    Diffusion Auto- Regressive Transformer for Effective Self-Supervised Time Series Forecasting

    Daoyu Wang, Mingyue Cheng, Zhiding Liu, Qi Liu, and Enhong Chen. Diffusion Auto- Regressive Transformer for Effective Self-Supervised Time Series Forecasting. arXiv preprint arXiv:2410.05711, 2024

    2024 arXiv

  46. [56]

    TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis

    Shiyu Wang, Jiawei Li, Xiaoming Shi, Zhou Ye, Baichuan Mo, Wenze Lin, Shengtong Ju, Zhixuan Chu, and Ming Jin. TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis. arXiv preprint arXiv:2410.16032, 2024

    arXiv

  47. [57]

    Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting

    Haixu Wu, Jiehui Xu, Jianmin Wang, and Mingsheng Long. Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting. In Neural Information Processing Systems, volume 34, pages 22419–22430, 2021

    2021

  48. [58]

    Interpretable Weather Forecasting for Worldwide Stations with a Unified Deep Model.Nature Machine Intelligence, 5(6):602–611, 2023

    Haixu Wu, Hang Zhou, Mingsheng Long, and Jianmin Wang. Interpretable Weather Forecasting for Worldwide Stations with a Unified Deep Model.Nature Machine Intelligence, 5(6):602–611, 2023

    2023

  49. [59]

    PromptCast: A New Prompt-Based Learning Paradigm for Time Series Forecasting

    Hao Xue and Flora D Salim. PromptCast: A New Prompt-Based Learning Paradigm for Time Series Forecasting. IEEE Transactions on Knowledge and Data Engineering, 36(11):6851–6864, 2023

    2023

  50. [60]

    Qwen2.5 Technical Report

    An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, et al. Qwen2.5 Technical Report. arXiv preprint arXiv:2412.15115, 2024

    2024 arXiv

  51. [61]

    TimeRAG: Boosting LLM Time Series Forecasting via Retrieval-Augmented Generation

    Silin Yang, Dong Wang, Haoqi Zheng, and Ruochun Jin. TimeRAG: Boosting LLM Time Series Forecasting via Retrieval-Augmented Generation. In ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE, 2025

    2025

  52. [62]

    Temporal Data Meets LLM–Explainable Financial Time Series Forecasting

    Xinli Yu, Zheng Chen, Yuan Ling, Shujing Dong, Zongyi Liu, and Yanbin Lu. Temporal Data Meets LLM–Explainable Financial Time Series Forecasting. arXiv preprint arXiv:2306.11025, 2023

    2023 arXiv

  53. [63]

    Diffusion-TS: Interpretable Diffusion for General Time Series Generation

    Xinyu Yuan and Yan Qiao. Diffusion-TS: Interpretable Diffusion for General Time Series Generation. arXiv preprint arXiv:2403.01742, 2024

    2024 arXiv

  54. [64]

    Are Transformers Effective for Time Series Forecasting? In AAAI Conference on Artificial Intelligence, volume 37, pages 11121–11128, 2023

    Ailing Zeng, Muxi Chen, Lei Zhang, and Qiang Xu. Are Transformers Effective for Time Series Forecasting? In AAAI Conference on Artificial Intelligence, volume 37, pages 11121–11128, 2023

    2023

  55. [65]

    How much can time-related features enhance time series forecasting? arXiv preprint arXiv:2412.01557, 2024

    Chaolv Zeng, Yuan Tian, Guanjie Zheng, and Yunjun Gao. How much can time-related features enhance time series forecasting? arXiv preprint arXiv:2412.01557, 2024. 16

    2024 arXiv

  56. [66]

    Solar Forecasting With Hourly Updated Numerical Weather Prediction

    Gang Zhang, Dazhi Yang, George Galanis, and Emmanouil Androulakis. Solar Forecasting With Hourly Updated Numerical Weather Prediction. Renewable and Sustainable Energy Reviews, 154:111768, 2022

    2022

  57. [67]

    Integration of Mamba and Transformer-MAT for Long-Short Range Time Series Forecasting With Application to Weather Dynamics

    Wenqing Zhang, Junming Huang, Ruotong Wang, Changsong Wei, Wenqian Huang, and Yuxin Qiao. Integration of Mamba and Transformer-MAT for Long-Short Range Time Series Forecasting With Application to Weather Dynamics. In 2024 International Conference on Electrical, Communication a...

    2024

  58. [68]

    Crossformer: Transformer Utilizing Cross-Dimension Depen- dency for Multivariate Time Series Forecasting

    Yunhao Zhang and Junchi Yan. Crossformer: Transformer Utilizing Cross-Dimension Depen- dency for Multivariate Time Series Forecasting. In The Eleventh International Conference on Learning Representations, 2023

    2023

  59. [69]

    Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting

    Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wancai Zhang. Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. In AAAI Conference on Artificial Intelligence, volume 35, pages 11106–11115, 2021

    2021

  60. [70]

    FiLM: Frequency Improved Legendre Memory Model for Long-Term Time Series Forecasting

    Tian Zhou, Ziqing Ma, Qingsong Wen, Liang Sun, Tao Yao, Wotao Yin, Rong Jin, et al. FiLM: Frequency Improved Legendre Memory Model for Long-Term Time Series Forecasting. In Neural Information Processing Systems, volume 35, pages 12677–12690, 2022

    2022

  61. [71]

    FEDformer: Fre- quency Enhanced Decomposed Transformer for Long-Term Series Forecasting

    Tian Zhou, Ziqing Ma, Qingsong Wen, Xue Wang, Liang Sun, and Rong Jin. FEDformer: Fre- quency Enhanced Decomposed Transformer for Long-Term Series Forecasting. InInternational Conference on Machine Learning, pages 27268–27286. PMLR, 2022

    2022

  62. [72]

    One Fits All: Power General Time Series Analysis by Pretrained LM

    Tian Zhou, Peisong Niu, Liang Sun, Rong Jin, et al. One Fits All: Power General Time Series Analysis by Pretrained LM. Advances in Neural Information Processing Systems, 36:43322– 43355, 2023

    2023

  63. [73]

    Unveil- ing the Potential of Text in High-Dimensional Time Series Forecasting

    Xin Zhou, Weiqing Wang, Shilin Qu, Zhiqiang Zhang, and Christoph Bergmeir. Unveil- ing the Potential of Text in High-Dimensional Time Series Forecasting. arXiv preprint arXiv:2501.07048, 2025

    2025 arXiv

  64. [74]

    Can LLMs Understand Time Series Anomalies? arXiv preprint arXiv:2410.05440, 2024

    Zihao Zhou and Rose Yu. Can LLMs Understand Time Series Anomalies? arXiv preprint arXiv:2410.05440, 2024. 17 A Timestamp Encoding The dataset utilized in this study incorporates three sampling frequencies: daily, weekly, and monthly. In the case of the daily dataset, it is rep...

    2021 arXiv

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.