VibeMathedMath problems solved by AI

The Lattice Triangle Problem in the Hard Obtuse Window

The lattice triangle problem asks which rational triangles unfold to Veech surfaces; in the hard obtuse window it is conjectured that none do. Via an arithmetic reformulation of the Mirzakhani-Wright rank obstruction, the paper rules out all but a density-0 subset of triangles in that window.

Result
Proved(see note)
Status
Partial result
AI contribution
AI-assisted
Method
Argument
Field
Teichmüller dynamics
Posed by
Year posed
Years open
Solved
2026-03-25
Model
AxiomProver
Vendor
Axiom Math
Collaborators
David Kurniadi Angdinata, Evan Chen, Ken Ono, Jiaxin Zhang, Jujian Zhang
Verification
Lean-checked, statement unaudited
Publication
Preprint
Significance
15 / 100
Disclosed cost
Wikipedia
No dedicated article

What was actually shown

A density-1 obstruction, not a full resolution: the conjecture that the hard window contains no lattice triangles remains open on a density-0 set.

What the AI did

"The main engine in this paper (Theorem 6.1) was autoformalized by AxiomProver in Lean (using mathlib)" - a test case for the autonomous system, with the protocol and artifacts documented in their own section.

Verification

The paper's main engine is autoformalized and kernel-checked in Lean by AxiomProver; the surrounding derivations are informal, and no independent review has appeared. Tier: the main engine was autoformalized by AxiomProver itself; the informal-to-formal correspondence is author-audited only.

Source

arXiv

Discussion