Yanasse: Finding New Proofs from Deep Vision’s Analogies Part 1: Probability Theory → \to Representation Theory
Alexandre Linhares · 2026
Read Yanasse: Finding New Proofs from Deep Vision’s Analogies Part 1: Probability Theory → \to Representation Theory by Alexandre Linhares (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-17229.txt.
Description
Linhares presents Yanasse, a method for finding new proofs by transferring proof patterns from distant mathematics. It extracts tactic usage across 27 Mathlib areas with 217,133 proof states. Scores flag tactics used in a source area but rare elsewhere. It matches proof states by analogy search, and asks an AI agent to adapt the pattern semantically, not by symbol substitution. For Probability to Representation Theory, it produces 4 verified proofs from 10 attempts, with zero sorries. Tactic schemas split into a gated head that rarely transfers and a general modifier.
Catalog record
| CREATOR | Alexandre Linhares |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42049 |
| LICENSE | http://creativecommons.org/licenses/by-sa/4.0/ |
| RIGHTS | CC BY SA |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.17229