KiT: A Foundation Model for Financial Time-Series Forecasting using DiffusionTransformers
cs.LG, cs.CV
Submitted: 2026-09-28
Updated: 2026-09-28
Code: https://github.com/Luciferbobo/KiT
License: http://creativecommons.org/licenses/by/4.0/
The gist: 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
Terminology
Abstract
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.
Sources
- Chronos-2: From Univariate to Universal Forecasting
- TimeDiT: General-purpose Diffusion Transformers for Time Series Foundation Model
- Classifier-Free Diffusion Guidance
- ELF: Embedded Language Flows
- FlowTS: Time Series Generation via Rectified Flow
- Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting
- Seedance 2.0: Advancing Video Generation for World Complexity
- GLU Variants Improve Transformer
- BloombergGPT: A Large Language Model for Finance
- FinGPT: Open-Source Financial Large Language Models
- TF-CoDiT: Conditional Time Series Synthesis with Diffusion Transformers for Treasury Futures
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks