New OptiDSL framework uses DSLs and LLMs for optimization modeling

Researchers have introduced OptiDSL, a new framework designed to improve the modeling of combinatorial optimization problems (COPs). Unlike existing systems that primarily use Mixed Integer Linear Programming (MILP), OptiDSL utilizes domain-specific languages (DSLs) and Large Language Models (LLMs) to decouple problem formulation from execution. This approach allows for integration with a wider variety of solvers, including heuristics and learning-based methods. Experiments on 44 COP types demonstrated that OptiDSL achieved a 51.66% gain in formulation accuracy and a 91.71% reduction in modeling time compared to MILP-based pipelines. AI

IMPACT This framework could accelerate research and development in optimization by making complex modeling more accessible and efficient.

RANK_REASON This is a research paper detailing a new framework for optimization modeling. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• combinatorial optimization problems
• Cops
• domain-specific language
• Large Language Models
• Mixed Integer Linear Programming
• OptiDSL

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →