LLM-QUBO: An End-to-End Framework for Automated QUBO Transformation from Natural Language Problem Descriptions

2025-08-27

AI Summary

Quantum annealing applications face two critical bottlenecks: manual, error-prone QUBO formulation and limited quantum hardware scale. Method: This paper proposes the first end-to-end optimization framework integrating large language models (LLMs) with quantum-classical co-decomposition. It leverages LLMs for semantic parsing of natural-language problems and automatic, correct QUBO generation; then applies Benders decomposition to partition the problem into a QUBO master problem (solved on quantum hardware) and a linear programming subproblem (solved classically). Contribution/Results: Empirical evaluation on classical platforms demonstrates 100% QUBO generation accuracy, significantly improved scalability via the hybrid strategy, and robust performance. The framework substantially lowers the barrier to adopting quantum optimization, while establishing a verifiable performance baseline and systematic methodology for deployment on real quantum hardware.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Satisfiability

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Quantum annealing offers a promising paradigm for solving NP-hard combinatorial optimization problems, but its practical application is severely hindered by two challenges: the complex, manual process of translating problem descriptions into the requisite Quadratic Unconstrained Binary Optimization (QUBO) format and the scalability limitations of current quantum hardware. To address these obstacles, we propose a novel end-to-end framework, LLM-QUBO, that automates this entire formulation-to-solution pipeline. Our system leverages a Large Language Model (LLM) to parse natural language, automatically generating a structured mathematical representation. To overcome hardware limitations, we integrate a hybrid quantum-classical Benders' decomposition method. This approach partitions the problem, compiling the combinatorial complex master problem into a compact QUBO format, while delegating linearly structured sub-problems to classical solvers. The correctness of the generated QUBO and the scalability of the hybrid approach are validated using classical solvers, establishing a robust performance baseline and demonstrating the framework's readiness for quantum hardware. Our primary contribution is a synergistic computing paradigm that bridges classical AI and quantum computing, addressing key challenges in the practical application of optimization problem. This automated workflow significantly reduces the barrier to entry, providing a viable pathway to transform quantum devices into accessible accelerators for large-scale, real-world optimization challenges.

Problem

• Automates QUBO transformation from natural language descriptions
• Overcomes quantum hardware scalability via hybrid decomposition
• Reduces entry barriers for practical quantum optimization applications

Innovation

• LLM automates natural language to QUBO transformation
• Hybrid quantum-classical Benders' decomposition enhances scalability
• Automated workflow bridges classical AI and quantum computing