Metraipolis › archive › dc_020b77cf2d7d
Learning to Optimize through Solver-Grounded Self-Play — arXiv:2609.34205
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34205
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:38:05.525Z
- SHA-256 of the archived text
82af0fb61702d7b22ff94d5be220b786cc83a2b305f2bedb6c71dc20bf44f240- Stable URL of this page
https://metraipolis.com/documento/dc_020b77cf2d7d
How to cite
Learning to Optimize through Solver-Grounded Self-Play — arXiv:2609.34205 — arXiv (en), https://arxiv.org/abs/2609.34205 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_020b77cf2d7d (Metraipolis archive dc_020b77cf2d7d, sha256 82af0fb61702d7b22ff94d5be220b786cc83a2b305f2bedb6c71dc20bf44f240)
BibTeX
@misc{dc020b77cf2d7d,
title = {Learning to Optimize through Solver-Grounded Self-Play — arXiv:2609.34205},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34205},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_020b77cf2d7d; sha256 82af0fb61702d7b22ff94d5be220b786cc83a2b305f2bedb6c71dc20bf44f240},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34205. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Optimization modeling is central to many decision-making scenarios, but traditionally requires extensive domain expertise. While Large Language Models (LLMs) have shown promise in automating this process, current training paradigms mainly rely on human-annotated or teacher-generated datasets. This dependence introduces a Generalization Ceiling, where models overfit to narrow data distributions, and Capability Anchoring, where models' reasoning is bounded by annotator proficiency and teacher model capability. In response, we propose OPT-Zero, the first fully self-play training framework for optimization modeling that requires zero external training data. OPT-Zero employs a single LLM in a dual-role closed loop: a Proposer that synthesizes increasingly challenging optimization problems alongside their mathematical formulations and solving code, and a Solver that attempts to resolve the problems given only natural-language problem descriptions. Grounded in execution feedback from external optimization solvers, we alternately train both roles using reinforcement learning. This process fosters an auto-curriculum in which the Proposer and Solver co-evolve: generating harder valid problems by the Proposer seamlessly enhances the structural reasoning ability of the Solver. Extensive results indicate that with zero curated data, OPT-Zero matches state-of-the-art data-dependent methods while exhibiting substantially stronger generalizability, establishing self-play training as a highly scalable paradigm for advancing LLM reasoning in modeling and solving optimization problems. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.34205 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T22:14:28.136Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).