SOTAVerified

Automated Theorem Proving

The goal of Automated Theorem Proving is to automatically generate a proof, given a conjecture (the target theorem) and a knowledge base of known facts, all expressed in a formal language. Automated Theorem Proving is useful in a wide range of applications, including the verification and synthesis of software and hardware systems.

Source: Learning to Prove Theorems by Learning to Generate Theorems

Papers

Showing 110 of 288 papers

TitleStatusHype
CriticLean: Critic-Guided Reinforcement Learning for Mathematical FormalizationCode1
Prover Agent: An Agent-based Framework for Formal Mathematical Proofs0
Towards Advanced Mathematical Reasoning for LLMs via First-Order Logic Theorem Proving0
MATP-BENCH: Can MLLM Be a Good Automated Theorem Prover for Multimodal Problems?0
Safe: Enhancing Mathematical Reasoning in Large Language Models via Retrospective Step-aware Formal VerificationCode1
LeanExplore: A search engine for Lean 4 declarationsCode2
Rewarding the Unlikely: Lifting GRPO Beyond Distribution Sharpening0
Faithful and Robust LLM-Driven Theorem Proving for NLI Explanations0
ProofNet++: A Neuro-Symbolic System for Formal Proof Verification with Self-Correction0
DeepTheorem: Advancing LLM Reasoning for Theorem Proving Through Natural Language and Reinforcement LearningCode1
Show:102550
← PrevPage 1 of 29Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Kimina-Prover-Previewcumulative80.74Unverified
2ProofAugcumulative66Unverified
3DeepSeek-Prover-V1.5cumulative63.5Unverified
4Subgoal-XLcumulative56.1Unverified
5DeepSeek-Provercumulative52Unverified
6Lyra + GPT-4cumulative47.1Unverified
7LEGO-Prover ChatGPTcumulative47.1Unverified
8Decomposing the Enigmacumulative45.5Unverified
9Evaristecumulative41Unverified
10Evariste-7dcumulative40.6Unverified
#ModelMetricClaimedVerifiedStatus
1EvaristePass@6458.6Unverified
2LEGO-Prover ChatGPTPass@10057Unverified
3Lyra + GPT-4Pass@10052Unverified
4Evariste-7dPass@6447.5Unverified
5GPT-fPass@6447.3Unverified
6Evariste-1dPass@6446.7Unverified
7DSP (62B Minerva informal)Pass@10043.9Unverified
8Lean GPT-fPass@829.3Unverified
9Lean tidyPass@116.8Unverified
10Metamath GPT-fPass@82Unverified
#ModelMetricClaimedVerifiedStatus
1MPNN-DagLSTMClassification Accuracy0.92Unverified
2FormulaNetClassification Accuracy0.9Unverified
3FormulaNet-basicClassification Accuracy0.89Unverified
4Siamese 1D CNN-LSTMClassification Accuracy0.83Unverified
5Siamese 1D CNNClassification Accuracy0.82Unverified
#ModelMetricClaimedVerifiedStatus
14-hop GNN, sub-expression sharingPercentage correct49.95Unverified
2Tactic Dependent LoopPercentage correct38.88Unverified
3BoW2 (extra -ves)Percentage correct36.55Unverified
4Deeper Wider WaveNetPercentage correct32.65Unverified
#ModelMetricClaimedVerifiedStatus
1FormulaNetClassification Accuracy0.9Unverified
2FormulaNet-basicClassification Accuracy0.89Unverified
31D CNN-LSTMClassification Accuracy0.83Unverified
41D CNNClassification Accuracy0.83Unverified
#ModelMetricClaimedVerifiedStatus
1EvaristePass@3272.4Unverified
2GPT-fPercentage correct56.2Unverified
3MetaGen-IL + HolophrasmPercentage correct22.1Unverified
4HolophrasmPercentage correct14.3Unverified
#ModelMetricClaimedVerifiedStatus
1Evariste-7dPass@6442.5Unverified
2Evariste-1dPass@6433.6Unverified
3EvaristePass@6432.1Unverified
4GPT-fPass@6430.6Unverified
#ModelMetricClaimedVerifiedStatus
1Proverbot9001Percentage correct19.36Unverified
2CoqGym/ASTacticPercentage correct4.99Unverified
#ModelMetricClaimedVerifiedStatus
1ASTacticPercentage correct12.2Unverified