Lean Refactor uses retrieval from a curated multi-objective strategy database to guide frozen LLMs in refactoring Lean proofs, reporting over 70% token compression on benchmarks and improved version transfer.
Seed-prover 1.5: Mastering undergraduate-level theorem proving via learning from experience, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Lean Refactor: Multi-Objective Controllable Proof Optimization via Agentic Strategy Search
Lean Refactor uses retrieval from a curated multi-objective strategy database to guide frozen LLMs in refactoring Lean proofs, reporting over 70% token compression on benchmarks and improved version transfer.