Large Language Models as Optimizers: A Survey of Direct vs. Tool-Augmented Approaches and Their Performance Frontiers
Proceedings of MIPRO 2026 — 49th ICT and Electronics Convention
Abstract
A survey of large language models used as optimizers, contrasting direct prompting approaches with tool-augmented pipelines and characterising the performance frontiers of each. (DOI/abstract pending verification.)
Notes
Survey of LLM-driven optimization, comparing direct and tool-augmented approaches. Metadata seeded from Google Scholar — verify DOI and abstract.
How to cite
@inproceedings{brcic2026brcic,
author = {Roko Peran and Luka Hobor and Mihael Kovac and Mario Brcic},
title = {Large Language Models as Optimizers: A Survey of Direct vs. Tool-Augmented Approaches and Their Performance Frontiers},
booktitle = {Proceedings of MIPRO 2026 — 49th ICT and Electronics Convention},
year = {2026},
}