MiniF2F: a cross-system benchmark for formal Olympiad-level mathematics

TL;DR

MiniF2F is a new cross-system benchmark dataset of Olympiad-level math problems designed to evaluate and advance neural theorem proving across multiple proof systems.

Contribution

It introduces miniF2F, a unified benchmark with problems from various math competitions, and provides baseline results using GPT-3 based neural theorem proving.

Findings

Baseline GPT-f achieves modest performance on miniF2F.

The benchmark covers diverse Olympiad-level problems from multiple proof systems.

MiniF2F aims to catalyze progress in neural theorem proving research.

Abstract

We present miniF2F, a dataset of formal Olympiad-level mathematics problems statements intended to provide a unified cross-system benchmark for neural theorem proving. The miniF2F benchmark currently targets Metamath, Lean, Isabelle (partially) and HOL Light (partially) and consists of 488 problem statements drawn from the AIME, AMC, and the International Mathematical Olympiad (IMO), as well as material from high-school and undergraduate mathematics courses. We report baseline results using GPT-f, a neural theorem prover based on GPT-3 and provide an analysis of its performance. We intend for miniF2F to be a community-driven effort and hope that our benchmark will help spur advances in neural theorem proving.