mlsanityData to 14 Oct 2026

KiT: A Foundation Model for Financial Time-Series Forecasting using DiffusionTransformers

arXiv 2609.34507 ↗First posted 28 Sep 2026cs.LG cs.CVcited by 0

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Financial candlestick forecasting is fundamental to quantitative investment, yet it remains exceptionally challenging due to extremely low signal-to-noise ratios and vast heterogeneity across markets and instruments. Existing approaches have largely attempted to introduce deep learning to capture hidden temporal features, but most adopt an auto-regressive formulation, which leads to error accumulation during inference. Meanwhile, general-purpose time-series foundation models are not tailored to the unique structure of k-line data and yield unsatisfactory performance on downstream candlestick forecasting tasks. To tackle these problems, we introduce KiT, a K-line Diffusion Transformer foundation model, and reformulate future prediction as conditional path generation via flow matching: given a historical context window, the model generates an ensemble of plausible future OHLCV trajectories. We pre-train KiT at multiple parameter scales on billions of candlestick bars spanning multiple markets and timescales. Across three markets and seven resolutions, KiT attains a mean return RankIC of 0.057 and a mean volatility RankIC of 0.66, leading at every timescale and outperforming both task-specific financial forecasters and general time-series foundation models. Code will be available at: https://github.com/Luciferbobo/KiT.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 57 of its 81 references that are in the map. First posted on arXiv 28 Sep 2026.

Builds on

57 of 81 references.

Closest ancestors · 13

Same direction, newest first

  1. 2026TF-CoDiT: Conditional Time Series Synthesis with Diffusion Transformers for Treasury Futures17 Jan
  2. 2025Non-stationary Diffusion For Probabilistic Time Series Forecasting7 May
  3. 2024FlowTS: Time Series Generation via Rectified Flow12 Nov
  4. 2024Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting3 Oct
  5. 2024TimeDiT: General-purpose Diffusion Transformers for Time Series Foundation Model3 Sep
  6. 2024DiffSTOCK: Probabilistic relational Stock Market Predictions using Diffusion Models21 Mar
  7. 2024Diffusion-TS: Interpretable Diffusion for General Time Series Generation4 Mar
  8. 2023Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series Forecasting21 Jul
  9. 2023Non-autoregressive Conditional Diffusion Models for Time Series Prediction8 Jun
  10. 2022Flow Matching for Generative Modeling6 Oct
  11. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow7 Sep
  12. 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation7 Jul
1 more
  1. 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting28 Jan

Other references · 44

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Denoising Diffusion Implicit Modelscited by 9,781
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  4. 2022Scalable Diffusion Models with Transformerscited by 6,054
  5. 2022Classifier-Free Diffusion Guidancecited by 5,847
  6. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  7. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  8. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  9. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
  10. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
  11. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
  12. 2019Root Mean Square Layer Normalizationcited by 1,480
32 more
  1. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,424
  2. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,210
  3. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,058
  4. 2023BloombergGPT: A Large Language Model for Financecited by 939
  5. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 931
  6. 2025Large Language Diffusion Modelscited by 912
  7. 2023PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesiscited by 864
  8. 2024Simple and Effective Masked Diffusion Language Modelscited by 848
  9. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 761
  10. 2024Chronos: Learning the Language of Time Seriescited by 759
  11. 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 700
  12. 2023A decoder-only foundation model for time-series forecastingcited by 692
  13. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 614
  14. 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 482
  15. 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 346
  16. 2020Query-Key Normalization for Transformerscited by 280
  17. 2023FinGPT: Open-Source Financial Large Language Modelscited by 253
  18. 2024Timer: Generative Pre-trained Transformers Are Large Time Series Modelscited by 208
  19. 2025Chronos-2: From Univariate to Universal Forecastingcited by 194
  20. 2025Sundial: A Family of Highly Capable Time Series Foundation Modelscited by 180
  21. 2023Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecastingcited by 113
  22. 2024Image and Video Tokenization with Binary Spherical Quantizationcited by 112
  23. 2022Why Exposure Bias Matters: An Imitation Learning Perspective of Error Accumulation in Language Generationcited by 108
  24. 2023Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglementcited by 90
  25. 2024FreDF: Learning to Forecast in the Frequency Domaincited by 79
  26. 2026ELF: Embedded Language Flowscited by 45
  27. 2024Scaling Law for Time Series Forecastingcited by 33
  28. 2024MarS: a Financial Market Simulation Engine Powered by Generative Foundation Modelcited by 28
  29. 2025Kronos: A Foundation Model for the Language of Financial Marketscited by 26
  30. 2024In-context Time Series Predictorcited by 23
  31. 2025FinCast: A Foundation Model for Financial Time-Series Forecastingcited by 10
  32. 2024LENS: Large Pre-trained Transformer for Exploring Financial Time Series Regularitiescited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Integrating Inductive Biases in Transformers via Distillation for Financial Time Series Forecasting
  2. 2024A Financial Time Series Denoiser Based on Diffusion Model
  3. 2025TRADES: Generating Realistic Market Simulations with Diffusion Models
  4. 2026Deep Learning for Financial Time Series: A Large-Scale Benchmark of Risk-Adjusted Performance
  5. 2024DiffsFormer: A Diffusion Transformer on Stock Factor Augmentation
  6. 2025FinZero: Launching Multi-modal Financial Time Series Forecast with Large Reasoning Model
  7. 2025Exploring Diffusion Models for Generative Forecasting of Financial Charts
  8. 2025Diffolio: A Diffusion Model for Multivariate Probabilistic Financial Time-Series Forecasting and Portfolio Construction