ReguSim: Evaluating LLM Agent Rule Grounding in Financial Compliance

Abstract

LLM agents in financial markets may cite rules yet still submit orders that violate executable constraints or misread surveillance evidence. We introduce ReguSim, a controlled financial-compliance environment, and ReguBench, a target-marked monitoring benchmark, to separate stated reasoning, attempted action, execution enforcement, and monitor evidence. Experiments show that visible rules reduce but do not eliminate rejected actions, incentives and persona framing shift behavior, and trader rationales can mislead an independent monitor unless enforcement evidence is available. The results frame financial-compliance evaluation as an audit of rule-grounded actions and evidence use rather than a single compliance score.