SITA: A Framework for Structure-to-Instance Theorem Autoformalization
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (23) · 19224-19232
DOI: 10.1609/aaai.v40i23.38997 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Chenyi Li und 3 weiteren Autor:innen.
Abstract (Originalsprache)
While large language models (LLMs) have shown progress in mathematical reasoning, they still face challenges in formalizing theorems that arise from instantiating abstract structures in concrete settings. With the goal of auto-formalizing mathematical results at the research level, we develop a framework for structure-to-instance theorem autoformalization (SITA), which systematically bridges the gap between abstract mathematical theories and their concrete applications in Lean proof assistant. Formalized abstract structures are treated as modular templates that contain definitions, assumptions, operations, and theorems. These templates serve as reusable guides for the formalization of concrete instances. Given a specific instantiation, we generate corresponding Lean definitions and instance declarations, integrate them using Lean’s typeclass mechanism, and construct verified theorems by checking structural assumptions. We incorporate LLM-based generation with feedback-guided refinement to ensure both automation and formal correctness. Experiments on a dataset of optimization problems demonstrate that SITA effectively formalizes diverse instances grounded in abstract structures.
Zitieren
APA
Li, C., Ma, W., Wang, Z., & Wen, Z. (2026). SITA: A Framework for Structure-to-Instance Theorem Autoformalization. Proceedings of the AAAI Conference on Artificial Intelligence, 40(23), 19224-19232. https://doi.org/10.1609/aaai.v40i23.38997
Vancouver
Li C, Ma W, Wang Z, Wen Z. SITA: A Framework for Structure-to-Instance Theorem Autoformalization. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(23):19224-19232. doi:10.1609/aaai.v40i23.38997
BibTeX
@article{Li2026,
author = {Li, Chenyi and Ma, Wanli and Wang, Zichen and Wen, Zaiwen},
title = {SITA: A Framework for Structure-to-Instance Theorem Autoformalization},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {23},
pages = {19224--19232},
doi = {10.1609/aaai.v40i23.38997}
}
EUVIMED