TheoremBench is a Lean4 benchmark of classical theorems in main and premised forms that evaluates LLM provers on partial progress, coverage, and token efficiency rather than binary success on competition problems.
arXiv preprint arXiv:2407.03203 , year=
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
abstract
Proving mathematical theorems using computer-verifiable formal languages like Lean significantly impacts mathematical reasoning. One approach to formal theorem proving involves generating complete proofs using Large Language Models (LLMs) based on Natural Language (NL) proofs. However, due to the scarcity of aligned NL and Formal Language (FL) theorem-proving data most modern LLMs exhibit suboptimal performance.This scarcity results in a paucity of methodologies for training LLMs and techniques to fully utilize their capabilities in composing formal proofs. To address these challenges, this paper proposes TheoremLlama, an end-to-end framework that trains a general-purpose LLM to be a Lean4 expert. TheoremLlama includes NL-FL dataset generation and bootstrapping method to obtain aligned dataset, curriculum learning and block training techniques to train the model, and iterative proof writing method to write Lean4 proofs that work together synergistically. Using the dataset generation method in TheoremLlama, we provide Open Bootstrapped Theorems (OBT), an NL-FL aligned and bootstrapped dataset. Our novel NL-FL bootstrapping method, where NL proofs are integrated into Lean4 code for training datasets, leverages the NL reasoning ability of LLMs for formal reasoning. The TheoremLlama framework achieves cumulative accuracies of 36.48% and 33.61% on MiniF2F-Valid and Test datasets respectively, surpassing the GPT-4 baseline of 22.95% and 25.41%. Our code, model checkpoints, and the generated dataset is published in GitHub
years
2026 6representative citing papers
Goedel-Architect introduces blueprint generation and iterative refinement for Lean 4 theorem proving, reaching 99.2% on MiniF2F-test and 75.6% on PutnamBench with DeepSeek-V4-Flash.
Retrieving structured thinking traces as a corpus improves reasoning performance on AIME, LiveCodeBench, and GPQA over standard RAG or no retrieval.
LiveFMBench shows that direct LLM prompting for C program formal specs overestimates accuracy by ~20% due to unfaithful behaviors like deceiving provers, while agentic workflows help under low sampling but overall performance remains far below human-authored specs.
LLM formal provers must shift from competition solvers to research agents that handle open-ended, under-specified frontier mathematics under machine-checked rigor.
citing papers explorer
-
TheoremBench: Evaluating LLMs on Theorem Proving in Formal Mathematics
TheoremBench is a Lean4 benchmark of classical theorems in main and premised forms that evaluates LLM provers on partial progress, coverage, and token efficiency rather than binary success on competition problems.
-
Goedel-Architect: Streamlining Formal Theorem Proving with Blueprint Generation and Refinement
Goedel-Architect introduces blueprint generation and iterative refinement for Lean 4 theorem proving, reaching 99.2% on MiniF2F-test and 75.6% on PutnamBench with DeepSeek-V4-Flash.
-
RAG over Thinking Traces Can Improve Reasoning Tasks
Retrieving structured thinking traces as a corpus improves reasoning performance on AIME, LiveCodeBench, and GPQA over standard RAG or no retrieval.
-
LiveFMBench: Unveiling the Power and Limits of Agentic Workflows in Specification Generation
LiveFMBench shows that direct LLM prompting for C program formal specs overestimates accuracy by ~20% due to unfaithful behaviors like deceiving provers, while agentic workflows help under low sampling but overall performance remains far below human-authored specs.
-
From Solvers to Research: Large Language Model-Driven Formal Mathematics at the Research Frontier
LLM formal provers must shift from competition solvers to research agents that handle open-ended, under-specified frontier mathematics under machine-checked rigor.
- Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics