aiqbench analysis
This research reveals a critical gap in current large language models, showing they can mimic valid limit order book sequences without truly understanding the underlying stochastic market dynamics. Consequently, relying on these models for financial forecasting introduces significant bias and spurious predictability that could mislead traders. The study underscores the necessity of rigorous world model testing before deploying LLMs in high-stakes quantitative environments.
Do LLMs Understand Limit Order Book Dynamics?
arXiv:2608.23706v1 Announce Type: new Abstract: A large language model (LLM) trained on synthetic limit order book (LOB) data achieves near perfect scores in generating valid sequences of LOB events. However, the LLM's implicit world model fails to learn the state of the LOB. This deficiency leads to biased estimates and spurious predictability in using the LLM to forecast future LOB events. Our analysis uses novel tests of an LLM's world model, extending prior work from deterministic settings to the stochastic dynamics needed for the LOB.
Title:Do LLMs Understand Limit Order Book Dynamics?
Abstract:A large language model (LLM) trained on synthetic limit order book (LOB) data achieves near perfect scores in generating valid sequences of LOB events. However, the LLM's implicit world model fails to learn the state of the LOB. This deficiency leads to biased estimates and spurious predictability in using the LLM to forecast future LOB events. Our analysis uses novel tests of an LLM's world model, extending prior work from deterministic settings to the stochastic dynamics needed for the LOB.
Artificial Intelligence (cs.AI)
arXiv:2608.23706 [cs.AI]
(or arXiv:2608.23706v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2608.23706 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 181 words
Related coverage
Presentation: Beyond Prompting: Context Engineering and Memory Management for AI Systems at Scale
InfoQ · Jun 10, 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
[AINews] Meta Muse Wonder & Wonder Full: MSL’s Claude 5.5-tier Frontier Model
Latent Space · Oct 13, 2026
Presentation: Enchant Your AI and APIs with eBPF Magic 🪄
InfoQ · Aug 21, 2026