Researchers teach LLMs to find interesting theorems, boosting interestingness 4.3x

CatAstro_Piyush · x · 2026-09-26

A new research thread argues that while LLMs can now prove results that resisted mathematicians for decades, finding interesting theorems without human guidance is the new bottleneck. The team introduces a quantitative notion of "interestingness," trains an LLM to optimize for it—achieving 4.3x higher interestingness—and builds a self-expanding discovery loop.

Original post →

More from Research

Research channel →