MINT: a Computer Program for Adaptive Monte Carlo Integration and Generation of Unweighted Distributions

TL;DR

The paper presents MINT, a memory-efficient, VEGAS-based tool for adaptive Monte Carlo integration and generation of unweighted distributions in multi-dimensional spaces. It introduces folding to handle non-positive integrands and uses a product-based upper-bound envelope to enable event generation without storing per-cell integrals, improving scalability over previous methods like SPRING-BASES. The approach yields accurate integration with adaptive grids and enables generation of p-tuples consistent with the target distribution, with direct applicability to NLO event generation such as POWHEG. The accompanying code provides a practical, public implementation with a flexible interface for folding and generation tasks.

Abstract

In this note I illustrate the program MINT, a FORTRAN program for Monte Carlo adaptive integration and generation of unweighted distributions.