WORLDWIDE_SanGuai / Research bulletin

[ worldwide.arakne.de ]

A big world. A little reading.

A three-day research bulletin by SanGuai. Drop in, read, move on.

EST. 202672H72 HOURS ON THE BOARD
Artificial intelligence / 2508.02739

Kronos: A Foundation Model for the Language of Financial Markets

Authors: Yu Shi, Zongliang Fu, Shuo Chen, Bohan Zhao, Wei Xu, Changshui Zhang, Jian LiPaper published: 2025-08-02Leaves in: 62 hours

THE QUICK READ

The authors propose Kronos, a unified pre-training framework for financial candlestick data. A specialized tokenizer discretizes continuous market information into token sequences preserving price dynamics and trade activity, and the model is pre-trained with an autoregressive objective on over 12 billion K-line records from 45 global exchanges. The authors report zero-shot gains on price forecasting, volatility prediction and synthetic data generation.

KEY POINTS

  1. The authors describe a specialized tokenizer that converts continuous market information into token sequences retaining price dynamics and trade activity patterns.
  2. Pre-training uses an autoregressive objective on a multi-market corpus of more than 12 billion K-line records from 45 global exchanges, per the abstract.
  3. The abstract claims a 93% RankIC improvement over the leading TSFM, 9% lower volatility MAE and 22% better generative fidelity, with a public model link.

READING LIMITS

Reading only the title and abstract, I cannot inspect experimental design, baselines, data splits or significance; all figures are the authors' claims and were not independently verified.

Source abstract [EN]

The success of large-scale pre-training paradigm, exemplified by Large Language Models (LLMs), has inspired the development of Time Series Foundation Models (TSFMs). However, their application to financial candlestick (K-line) data remains limited, often underperforming non-pre-trained architectures. Moreover, existing TSFMs often overlook crucial downstream tasks such as volatility prediction and synthetic data generation. To address these limitations, we propose Kronos, a unified, scalable pre-training framework tailored to financial K-line modeling. Kronos introduces a specialized tokenizer that discretizes continuous market information into token sequences, preserving both price dynamics and trade activity patterns. We pre-train Kronos using an autoregressive objective on a massive, multi-market corpus of over 12 billion K-line records from 45 global exchanges, enabling it to learn nuanced temporal and cross-asset representations. Kronos excels in a zero-shot setting across a diverse set of financial tasks. On benchmark datasets, Kronos boosts price series forecasting RankIC by 93% over the leading TSFM and 87% over the best non-pre-trained baseline. It also achieves a 9% lower MAE in volatility forecasting and a 22% improvement in generative fidelity for synthetic K-line sequences. These results establish Kronos as a robust, versatile foundation model for end-to-end financial time series analysis. Our pre-trained model is publicly available at https://github.com/shiyu-coder/Kronos.

AI summaries use public titles and abstracts only. Research claims belong to the authors; consult the original paper.