Reproduce “FlowLOB: Efficient and Controllable Limit Order Book Generation with Flow Matching” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Zhuohan Wang, Andreea Bacalum, Ollie Olby, Carmine Ventre, Namid Stillman

FlowLOB generates conditional continuations of limit order books from a book and requested market scenario. It uses flow matching with tick-relative prices and volumes, and a setup trained separately at each sampling frequency. In matched tests, 10-step Euler sampling achieved 4–120× lower Wasserstein-1 distance than diffusion across price and volume marginals; results concern network calls, with realism weaker at 10 seconds.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.