Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2507.08665 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916838963675136 |
|---|---|
| author | Zhang, Jiyao Zhong, Chengli Xu, Hui Li, Qige Zhou, Yi |
| author_facet | Zhang, Jiyao Zhong, Chengli Xu, Hui Li, Qige Zhou, Yi |
| contents | Modern large language models (LLMs) show promising progress in formalizing informal mathematics into machine-verifiable theorems. However, these methods still face bottlenecks due to the limited quantity and quality of multilingual parallel corpora. In this paper, we propose a novel neuro-symbolic framework KELPS (Knowledge-Equation based Logical Processing System) to address these problems. KELPS is an iterative framework for translating, synthesizing, and filtering informal data into multiple formal languages (Lean, Coq, and Isabelle). First, we translate natural language into Knowledge Equations (KEs), a novel language that we designed, theoretically grounded in assertional logic. Next, we convert them to target languages through rigorously defined rules that preserve both syntactic structure and semantic meaning. This process yielded a parallel corpus of over 60,000 problems. Our framework achieves 88.9% syntactic accuracy (pass@1) on MiniF2F, outperforming SOTA models such as Deepseek-V3 (81%) and Herald (81.3%) across multiple datasets. All datasets and codes are available in the supplementary materials. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_08665 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | KELPS: A Framework for Verified Multi-Language Autoformalization via Semantic-Syntactic Alignment Zhang, Jiyao Zhong, Chengli Xu, Hui Li, Qige Zhou, Yi Computation and Language Artificial Intelligence Modern large language models (LLMs) show promising progress in formalizing informal mathematics into machine-verifiable theorems. However, these methods still face bottlenecks due to the limited quantity and quality of multilingual parallel corpora. In this paper, we propose a novel neuro-symbolic framework KELPS (Knowledge-Equation based Logical Processing System) to address these problems. KELPS is an iterative framework for translating, synthesizing, and filtering informal data into multiple formal languages (Lean, Coq, and Isabelle). First, we translate natural language into Knowledge Equations (KEs), a novel language that we designed, theoretically grounded in assertional logic. Next, we convert them to target languages through rigorously defined rules that preserve both syntactic structure and semantic meaning. This process yielded a parallel corpus of over 60,000 problems. Our framework achieves 88.9% syntactic accuracy (pass@1) on MiniF2F, outperforming SOTA models such as Deepseek-V3 (81%) and Herald (81.3%) across multiple datasets. All datasets and codes are available in the supplementary materials. |
| title | KELPS: A Framework for Verified Multi-Language Autoformalization via Semantic-Syntactic Alignment |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2507.08665 |