WeSearch

Euclean: Automated Geometry Problem Formalization with Unified Verification in Lean

·3 min read · 0 reactions · 0 comments · 9 views
#euclean#automated#geometry#problem#formalization
Euclean: Automated Geometry Problem Formalization with Unified Verification in Lean
TL;DR · WeSearch summary

This split increases the trusted computing base and hinders unified model development. Existing geometry-in-Lean efforts (LeanEuclid, LeanGeo) introduce custom axiom systems incompatible with standard Mathlib, and their small scale ($<$ 1,100 problems) limits large-scale training. Native Mathlib autoformalization of geometry, however, poses distinct challenges: implicit diagrammatic assumptions (e.g., topological configuration and non-degeneracy) must be made explicit rather than deferred to external solvers, and models must adapt to Mathlib's small, rapidly evolving geometry infrastructure.

Key facts
Original article
arXiv.org
Read full at arXiv.org →
Opening excerpt (first ~120 words) tap to expand

Computer Science > Artificial Intelligence arXiv:2607.19374 (cs) [Submitted on 17 Jun 2026] Title:Euclean: Automated Geometry Problem Formalization with Unified Verification in Lean Authors:Linbin Tang, Jingyan You, Zilin Kang, Hanzhang Liu, Sophia Zhang, Zenan Li, Chenrui Cao, Liangcheng Song, Jiaao Wu, Xian Zhang, Fan Yang View a PDF of the paper titled Euclean: Automated Geometry Problem Formalization with Unified Verification in Lean, by Linbin Tang and 10 other authors View PDF HTML (experimental) Abstract:Recent formal reasoning systems have reached IMO-level performance, yet they leave a fragmented landscape: algebra and number theory are handled in Lean, while geometry still relies on domain-specific languages with limited formal guarantees.

Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.

Anonymous · no account needed
Share 𝕏 Facebook Reddit LinkedIn Threads WhatsApp Bluesky Mastodon Email

Discussion

0 comments

More from arXiv.org