$\mathtt{\text{ReMind}}$: Understanding Deductive Code Reasoning in LLMs

Large Language Models (LLMs) have achieved remarkable progress in code-related tasks. Despite their advancement, empirical evidence reveals that they still struggle with \emph{deductive code reasoning}, the ability to reason about the program execution process. While prior studies have recognized this limitation, the underlying causes remain largely underexplored. In this paper, we begin by presenting a comprehensive empirical study that reveals three key challenges undermining deductive code reasoning: (1) an intrinsic gap between generation and reasoning abilities, (2) a consistent bias towards code sources, and (3) weak zero-shot generalization on complex benchmarks. In light of these challenges, we propose $\mathtt{\text{ReMind}}$, a multi-agent framework composed of $\mathtt{\text{Mutator}}$, $\mathtt{\text{Executor}}$, and $\mathtt{\text{Inspector}}$. The $\mathtt{\text{Mutator}}$ generates code variants to mitigate bias towards code sources, the $\mathtt{\text{Executor}}$ traces variable states step-by-step to expose inconsistency, and the $\mathtt{\text{Inspector}}$ identifies problematic reasoning steps and provides control-flow refinement to bridge the intrinsic reasoning gap. Through their coordinated collaboration, $\mathtt{\text{ReMind}}$ systematically identifies and refines reasoning flaws, achieving outstanding performance and enabling robust zero-shot generalization. Extensive experiments on two benchmarks with five LLMs demonstrate the superior advantages of $\mathtt{\text{ReMind}}$ compared to baseline approaches in deductive code reasoning.