LEVER: Adaptive Cost-Aware Proof Search Over AND/OR Graphs
Quick summary
arXiv:2610.11862v1 Announce Type: new Abstract: Mathematicians value proofs for more than correctness: among correct proofs, simplicity, purity and the computational cost of finding them vary widely. Yet LLM-powered theorem provers largely search for any correct proof, and improve its quality only after it is found. We propose LEVER, a proof search algorithm that makes the objective over correct proofs programmable and optimizes it during search. LEVER scores partial proofs over an AND/OR proof graph, combining realized objective values with predictions for open subgoals, so the objective guid
Key takeaways
- arXiv:2610.11862v1 Announce Type: new Abstract: Mathematicians value proofs for more than correctness: among correct proofs, simplicity, purity and the computational cost of finding them vary widely.
- Yet LLM-powered theorem provers largely search for any correct proof, and improve its quality only after it is found.
- We propose LEVER, a proof search algorithm that makes the objective over correct proofs programmable and optimizes it during search.
Why it matters
The importance of “LEVER: Adaptive Cost-Aware Proof Search Over AND/OR Graphs” will be measured by what changes in practice. User behavior, access conditions, verifiable performance and responsible-use outcomes are the signals worth following.
Member comments