Automated Reformulation of Robust Optimization via Memory-Augmented Large Language Models

TL;DR

This paper introduces AutoRO-Bench, a benchmark for evaluating LLM-based robust optimization reformulation, and AutoREM, a memory-augmented framework that improves reformulation accuracy without domain expertise or parameter tuning.

Contribution

It presents a new benchmark for systematic evaluation of LLM-based RO reformulation and a novel memory-augmented method that enhances reformulation performance across models and datasets.

Findings

AutoREM improves reformulation accuracy across datasets.

AutoREM enhances efficiency without domain-specific knowledge.

AutoRO-Bench enables systematic evaluation of LLM-based reformulation.

Abstract

Robust optimization (RO) provides a principled framework for decision-making under uncertainty, but its practical use is often limited by the need to manually reformulate uncertain optimization models into tractable deterministic counterparts. Recent large language models (LLMs) have been shown promising for automating optimization formulation, yet RO reformulation remains challenging because it requires precise multi-step reasoning and mathematically consistent transformations. To facilitate systematic evaluation of LLM-based reformulation, for which no dedicated benchmark currently exists, we develop AutoRO-Bench, a benchmark featuring an automated data generation pipeline for the core RO reformulation task and a curated dataset for the RO application task. To address the reformulation challenge, we propose Automated Reformulation with Experience Memory (AutoREM), a tuning-free memory-augmented framework that autonomously builds a structured textual experience memory by reflecting on past failed trajectories through a tailored offline adaptation procedure. AutoREM requires neither domain-specific expert knowledge nor parameter updates, and the resulting memory readily transfers across different base LLMs. Experimental results show that AutoREM consistently improves the accuracy and efficiency of RO reformulation across in-distribution datasets, out-of-distribution datasets, and diverse base LLMs.