Modeling Copilots for
Text-to-Model Translation

A comprehensive suite of LLM-based modeling copilots for text-to-model translation. Includes multiple strategies of varying complexity — from zero-shot and chain-of-thought prompting to compositional and agentic workflows — along with an online leaderboard for benchmarking and an interactive MiniZinc editor with a built-in AI assistant.

Serdar Kadıoğlu, Karthik Uppuluri, Akash Singirikonda

@misc{kadioglu2026text2model,
  title         = {Modeling Copilots for Text-to-Model Translation},
  author        = {Serdar Kadioglu and Karthik Uppuluri and Akash Singirikonda},
  year          = {2026},
  eprint        = {2604.12955},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  url           = {https://arxiv.org/abs/2604.12955}
}

Fine-tuning state-of-the-art open-source LLMs to generate MiniZinc constraint models directly from natural language problem descriptions. Includes training datasets with plain and chain-of-thought variants, augmented data, and models spanning multiple architectures and sizes — from 0.6B to 20B parameters.

Serdar Kadıoğlu, Karthik Uppuluri, Akash Singirikonda

@misc{kadioglu2026text2model,
  title         = {Modeling Copilots for Text-to-Model Translation},
  author        = {Serdar Kadioglu and Karthik Uppuluri and Akash Singirikonda},
  year          = {2026},
  eprint        = {2604.12955},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  url           = {https://arxiv.org/abs/2604.12955}
}

A cross-domain dataset capturing optimization and satisfaction problems specified in natural language text. Leverages MiniZinc's solver-agnostic modeling capabilities to formulate a diverse range of combinatorial problems, bridging natural language specifications with formal constraint models. The first dataset to integrate both satisfaction and optimization problems within a unified framework.

Akash Singirikonda, Serdar Kadıoğlu, Karthik Uppuluri

@misc{singirikonda2025text2zinc,
  title         = {Text2Zinc: A Cross-Domain Dataset for Modeling Optimization and Satisfaction Problems in MiniZinc},
  author        = {Akash Singirikonda and Serdar Kadioglu and Karthik Uppuluri},
  year          = {2025},
  eprint        = {2503.10642},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CL},
  url           = {https://arxiv.org/abs/2503.10642}
}

Global LLM Agents for text-to-model translation. Introduces collaborative multi-agent workflows where specialized LLM agents work together to decompose, formulate, and validate optimization models from natural language — achieving stronger performance through agent collaboration.

Junyang Cai, Serdar Kadıoğlu, Bistra Dilkina

@misc{cai2025gala,
  title         = {Gala: Global LLM Agents for Text-to-Model Translation},
  author        = {Junyang Cai and Serdar Kadioglu and Bistra Dilkina},
  year          = {2025},
  eprint        = {2509.08970},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  url           = {https://arxiv.org/abs/2509.08970}
}

Named entity recognition for optimization modeling from natural language. Extracts optimization-related entities — objectives, decision variables, constraints, and parameters — from free-form problem descriptions. Combines classical NLP techniques with fine-tuned transformer architectures and optimization-specific data augmentation.

Serdar Kadıoğlu, Parag Pravin Dakle, Karthik Uppuluri, Regina Politi, Preethi Raghavan, SaiKrishna Rallabandi, Ravisutha Srinivasamurthy

@article{Kadioglu24,
  author    = {Serdar Kadioglu and Parag Pravin Dakle and Karthik Uppuluri and Regina Politi and Preethi Raghavan and SaiKrishna Rallabandi and Ravisutha Srinivasamurthy},
  title     = {Ner4Opt: named entity recognition for optimization modelling from natural language},
  journal   = {Constraints},
  volume    = {29},
  number    = {3-4},
  pages     = {261--299},
  year      = {2024},
  url       = {https://doi.org/10.1007/s10601-024-09376-5},
  doi       = {10.1007/S10601-024-09376-5}
}

@inproceedings{dakle23,
  title     = {Ner4Opt: Named Entity Recognition for Optimization Modelling from Natural Language},
  author    = {Parag Pravin Dakle and Serdar Kadıoğlu and Karthik Uppuluri and Regina Politi and Preethi Raghavan and SaiKrishna Rallabandi and Ravisutha Srinivasamurthy},
  booktitle = {The 20th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research (CPAIOR 2023)},
  year      = {2023}
}

@inproceedings{ner4opt2022neurips,
  title     = {Ner4Opt: Named Entity Recognition for Optimization Modelling from Natural Language},
  author    = {Parag Pravin Dakle and Serdar Kadıoğlu and Karthik Uppuluri and Regina Politi and Preethi Raghavan and SaiKrishna Rallabandi and Ravisutha Srinivasamurthy},
  booktitle = {NeurIPS 2022 Workshop on Optimization for Machine Learning (OPT)},
  year      = {2022}
}