Growing an Agent/Prover Interface: Evolutionary Tool Design for Cost-Efficient Theorem Proving in Rocq and Lean
What it is
This paper proposes an evolutionary method to design optimized Model Context Protocol (MCP) servers for theorem provers like Rocq and Lean. It uses a frontier model to incrementally propose and test new interface features, retaining only those that improve the performance of smaller agent models in terms of success rate, cost, and time.
The problem it solves
Standard interfaces for theorem provers are designed for human users and are not optimized for the specific interaction patterns and cost constraints of AI agents, leading to inefficient and expensive proof generation.
What you could use it for
- Automated formal verification of mathematical proofs
- Reducing the computational cost of LLM-based theorem proving
- Optimizing agent-prover interaction interfaces for specific mathematical domains
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.