Guardado en:
Detalles Bibliográficos
Autores principales: Song, Yang, Vyas, Anoushka, Wei, Zirui, Pakazad, Sina Khoshfetrat, Ohlsson, Henrik, Neubig, Graham
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2601.21372
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913170899075072
author Song, Yang
Vyas, Anoushka
Wei, Zirui
Pakazad, Sina Khoshfetrat
Ohlsson, Henrik
Neubig, Graham
author_facet Song, Yang
Vyas, Anoushka
Wei, Zirui
Pakazad, Sina Khoshfetrat
Ohlsson, Henrik
Neubig, Graham
contents We present NEMO, a system that translates Natural-language descriptions of decision problems into formal Executable Mathematical Optimization implementations using autonomous coding agents (ACAs). Existing approaches rely on specialized large language models (LLMs) or bespoke task-specific agents that are often brittle and frequently generate syntactically invalid or non-executable code. NEMO instead treats ACAs as a first-class abstraction analogous to API-based interaction with LLMs; their sandboxed execution guarantees code is executable by construction and supports automated validation and repair. We introduce novel coordination patterns including asymmetric validation loops between independently generated optimizer and simulator implementations, external memory for experience reuse, and robustness enhancements via minimum Bayes risk (MBR) decoding and self-consistency. Across nine established optimization benchmarks, NEMO achieves state-of-the-art performance on the majority of tasks with substantial margins on several datasets, demonstrating the power of execution-aware agentic architectures for automated optimization modeling.
format Preprint
id arxiv_https___arxiv_org_abs_2601_21372
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle NEMO: Execution-Aware Optimization Modeling via Autonomous Coding Agents
Song, Yang
Vyas, Anoushka
Wei, Zirui
Pakazad, Sina Khoshfetrat
Ohlsson, Henrik
Neubig, Graham
Artificial Intelligence
We present NEMO, a system that translates Natural-language descriptions of decision problems into formal Executable Mathematical Optimization implementations using autonomous coding agents (ACAs). Existing approaches rely on specialized large language models (LLMs) or bespoke task-specific agents that are often brittle and frequently generate syntactically invalid or non-executable code. NEMO instead treats ACAs as a first-class abstraction analogous to API-based interaction with LLMs; their sandboxed execution guarantees code is executable by construction and supports automated validation and repair. We introduce novel coordination patterns including asymmetric validation loops between independently generated optimizer and simulator implementations, external memory for experience reuse, and robustness enhancements via minimum Bayes risk (MBR) decoding and self-consistency. Across nine established optimization benchmarks, NEMO achieves state-of-the-art performance on the majority of tasks with substantial margins on several datasets, demonstrating the power of execution-aware agentic architectures for automated optimization modeling.
title NEMO: Execution-Aware Optimization Modeling via Autonomous Coding Agents
topic Artificial Intelligence
url https://arxiv.org/abs/2601.21372