Lean4Agent: Formal Modeling and Verification for Agent Workflow and Trajectory
Ruida Wang, Jerry Huang, Pengcheng Wang, Xuanqing Liu, Luyang Kong, Tong Zhang
Why It Matters
What makes this one worth your time
Formal verification of agent workflows can enhance the reliability and robustness of AI systems, which is crucial for deploying them in real-world applications.
Lean4Agent uses formal methods to improve the reliability of agent workflows.
Summary
The paper introduces Lean4Agent, a framework using Lean4, a dependent-type formal language, to model and verify agent workflows and execution trajectories. It includes FormalAgentLib, a library for formal modeling and verification, and LeanEvolve, which enhances workflows based on verification results. Experiments show improved performance in verified workflows and further enhancements with LeanEvolve.
Key contributions
- Introduction of Lean4Agent framework for formal modeling and verification of agent workflows.
- Development of FormalAgentLib, an extensible library for formal verification.
- Creation of LeanEvolve to enhance workflow capabilities based on verification results.
Notable insights
- Using a dependent-type formal language like Lean4 for agent modeling can help ensure semantic consistency and localize execution-time failures.
Possible limitations
- Not stated in the abstract
Abstract
arXiv:2606.06523v1 Announce Type: new Abstract: Equipping Large Language Models (LLMs) to execute reliable multi-step workflows has become a central challenge in artificial intelligence. Despite recent advances in LLMs' agentic capabilities, most agent systems still lack formal methods for specifying, verifying, and debugging their workflow and execution trajectories. This challenge mirrors a long-standing problem in mathematics, where the ambiguity of natural languages (NLs) motivates the development of formal languages (FLs). Inspired by this paradigm, we propose **Lean4Agent**, to the best of our knowledge, the first framework that uses Lean4, a dependent-type FL to model and verify agent behavior. **Lean4Agent** launches **FormalAgentLib**, an extensible Lean4 library for formally modeling and verifying agent workflows' semantic consistency under explicit assumptions, and enabling localization of execution-time failures revealed by trajectories. Building on **FormalAgentLib**, we further develop **LeanEvolve**, which applies results in **FormalAgentLib** to revise workflows to enhance its capability. Extensive experiments on a hard problem subset of SWE-Bench-Verified and a subset of ELAIP-Bench across 5 leading LLMs indicate that the verification-passing workflows outperform the failing ones by an average of **11.94%**, and **LeanEvolve** further improves SWE performance by **7.47%** on average. Furthermore, **Lean4Agent** establishes a foundation for a new field of using expressive dependent-type FL to formally model and verify agent behavior.