LRM-1B: Towards Large Routing Model

TL;DR

This paper introduces LRM-1B, a large neural routing model with 1 billion parameters, demonstrating state-of-the-art performance across various vehicle routing problem variants and analyzing the effects of model scaling.

Contribution

The paper presents the first large-scale neural routing model with 1 billion parameters and provides a comprehensive evaluation and analysis of model scaling effects in VRPs.

Findings

LRM-1B achieves state-of-the-art results on multiple VRP variants.

Model scaling from 1M to 1B parameters follows a power-law performance trend.

LRM-1B adapts effectively to diverse VRP scenarios.

Abstract

Vehicle routing problems (VRPs) are central to combinatorial optimization with significant practical implications. Recent advancements in neural combinatorial optimization (NCO) have demonstrated promising results by leveraging neural networks to solve VRPs, yet the exploration of model scaling within this domain remains underexplored. Inspired by the success of model scaling in large language models (LLMs), this study introduces a Large Routing Model with 1 billion parameters (LRM-1B), designed to address diverse VRP scenarios. We present a comprehensive evaluation of LRM-1B across multiple problem variants, distributions, and sizes, establishing state-of-the-art results. Our findings reveal that LRM-1B not only adapts to different VRP challenges but also showcases superior performance, outperforming existing models. Additionally, we explore the scaling behavior of neural routing models from 1M to 1B parameters. Our analysis confirms power-law between multiple model factors and performance, offering critical insights into the optimal configurations for foundation neural routing solvers.