REVIEW 8 cited by
LEGO-Prover: Neural Theorem Proving with Growing Libraries
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Despite the success of large language models (LLMs), the task of theorem proving still remains one of the hardest reasoning tasks that is far from being fully solved. Prior methods using language models have demonstrated promising results, but they still struggle to prove even middle school level theorems. One common limitation of these methods is that they assume a fixed theorem library during the whole theorem proving process. However, as we all know, creating new useful theorems or even new theories is not only helpful but crucial and necessary for advancing mathematics and proving harder and deeper results. In this work, we present LEGO-Prover, which employs a growing skill library containing verified lemmas as skills to augment the capability of LLMs used in theorem proving. By constructing the proof modularly, LEGO-Prover enables LLMs to utilize existing skills retrieved from the library and to create new skills during the proving process. These skills are further evolved (by prompting an LLM) to enrich the library on another scale. Modular and reusable skills are constantly added to the library to enable tackling increasingly intricate mathematical problems. Moreover, the learned library further bridges the gap between human proofs and formal proofs by making it easier to impute missing steps. LEGO-Prover advances the state-of-the-art pass rate on miniF2F-valid (48.0% to 57.0%) and miniF2F-test (45.5% to 47.1%). During the proving process, LEGO-Prover also manages to generate over 20,000 skills (theorems/lemmas) and adds them to the growing library. Our ablation study indicates that these newly added skills are indeed helpful for proving theorems, resulting in an improvement from a success rate of 47.1% to 50.4%. We also release our code and all the generated skills.
Forward citations
Cited by 8 Pith papers
-
StepFun-Prover Preview: Let's Think and Verify Step by Step
A reinforcement learning pipeline with Lean verifier feedback trains a 32B model that scores 70.0% pass@1 on miniF2F-test, a new state of the art.
-
Mathesis: Towards Formal Theorem Proving from Natural Languages
An RL-trained autoformalizer plus a Lean prover solves 18% of Chinese Gaokao proof problems end-to-end from natural language, and 64.3% of MiniF2F at pass@32.
-
Towards Generating Controllable and Solvable Geometry Problem by Leveraging Symbolic Deduction Engine
SDE-GPG samples from a knowledge-point-to-definition mapping table, runs the AlphaGeometry symbolic deduction engine to produce conclusions, filters candidates with a checking function, and translates the formal outpu...
-
Theory-Level Autoformalization: From Isolated Statements to Unified Formal Knowledge Bases
Autoformalization should target complete theories—axioms, definitions, lemmas, proofs—as unified formal libraries, not isolated statements.
-
Solving Formal Math Problems by Decomposition and Iterative Reflection
An agent that decomposes Lean 4 goals into subproblems and iteratively repairs proofs achieves a 95.9% pass rate on miniF2F-test using a stock Gemini model.
-
A Compute-Matched Re-Evaluation of TroVE on MATH
After matching computational budget, TroVE's toolbox mechanism yields only a marginal, statistically non-significant 1% accuracy gain over a plain sampling baseline on MATH.
-
Clarifying Before Reasoning: A Coq Prover with Structural Context
Enriching LLM theorem-proving prompts with Coq's internal type representations and natural-language explanations raises proof success from 21.8% to 45.8%, surpassing Graph2Tac's 33.2%.
-
StepProof: Step-by-step verification of natural language mathematical proofs
Decomposing natural-language proofs into sentence-level formal subproofs improves autoformalization success rates and efficiency compared with whole-proof formalization.
Discussion (0). Continue with ORCID to comment.