Research

Paper

TESTING February 27, 2026

TradeFM: A Generative Foundation Model for Trade-flow and Market Microstructure

Authors

Maxime Kawawa-Beaudan, Srijan Sood, Kassiani Papasotiriou, Daniel Borrajo, Manuela Veloso

Abstract

Foundation models have transformed domains from language to genomics by learning general-purpose representations from large-scale, heterogeneous data. We introduce TradeFM, a 524M-parameter generative Transformer that brings this paradigm to market microstructure, learning directly from billions of trade events across >9K equities. To enable cross-asset generalization, we develop scale-invariant features and a universal tokenization scheme that map the heterogeneous, multi-modal event stream of order flow into a unified discrete sequence -- eliminating asset-specific calibration. Integrated with a deterministic market simulator, TradeFM-generated rollouts reproduce key stylized facts of financial returns, including heavy tails, volatility clustering, and absence of return autocorrelation. Quantitatively, TradeFM achieves 2-3x lower distributional error than Compound Hawkes baselines and generalizes zero-shot to geographically out-of-distribution APAC markets with moderate perplexity degradation. Together, these results suggest that scale-invariant trade representations capture transferable structure in market microstructure, opening a path toward synthetic data generation, stress testing, and learning-based trading agents.

Metadata

arXiv ID: 2602.23784
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-02-27
Fetched: 2026-03-02 06:04

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.23784v1</id>\n    <title>TradeFM: A Generative Foundation Model for Trade-flow and Market Microstructure</title>\n    <updated>2026-02-27T08:26:53Z</updated>\n    <link href='https://arxiv.org/abs/2602.23784v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.23784v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Foundation models have transformed domains from language to genomics by learning general-purpose representations from large-scale, heterogeneous data. We introduce TradeFM, a 524M-parameter generative Transformer that brings this paradigm to market microstructure, learning directly from billions of trade events across &gt;9K equities. To enable cross-asset generalization, we develop scale-invariant features and a universal tokenization scheme that map the heterogeneous, multi-modal event stream of order flow into a unified discrete sequence -- eliminating asset-specific calibration. Integrated with a deterministic market simulator, TradeFM-generated rollouts reproduce key stylized facts of financial returns, including heavy tails, volatility clustering, and absence of return autocorrelation. Quantitatively, TradeFM achieves 2-3x lower distributional error than Compound Hawkes baselines and generalizes zero-shot to geographically out-of-distribution APAC markets with moderate perplexity degradation. Together, these results suggest that scale-invariant trade representations capture transferable structure in market microstructure, opening a path toward synthetic data generation, stress testing, and learning-based trading agents.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='q-fin.CP'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='q-fin.TR'/>\n    <published>2026-02-27T08:26:53Z</published>\n    <arxiv:comment>29 pages, 17 figures, 6 tables. Preprint</arxiv:comment>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Maxime Kawawa-Beaudan</name>\n    </author>\n    <author>\n      <name>Srijan Sood</name>\n    </author>\n    <author>\n      <name>Kassiani Papasotiriou</name>\n    </author>\n    <author>\n      <name>Daniel Borrajo</name>\n    </author>\n    <author>\n      <name>Manuela Veloso</name>\n    </author>\n  </entry>"
}