aiqbench analysis
This research introduces TACIT, a framework that leverages Large Language Models to automatically repair misspecified optimization models by integrating their reasoning with formal mathematical constraints. By combining top-down structural proposals from LLMs with bottom-up inference from historical data, the system significantly improves model accuracy in real-world scenarios where expert tacit knowledge is difficult to formalize. The approach demonstrates a notable reduction in overfitting compared to traditional inverse optimization methods, offering a practical path toward more robust decision-making systems.
TACIT: Optimization Models that Learn from Their Mistakes
arXiv:2609.38434v1 Announce Type: cross Abstract: Real-world optimization problems are difficult to model accurately because many objectives and constraints reside in domain experts' tacit knowledge, making them hard to formalize. As a result, optimization models often contain miscalibrated objectives, missing constraints, or omitted decision variables, leading to solutions that fail to reflect operational realities. We address this challenge by automatically repairing misspecified formulations using historical data consisting of past solutions and subsequent user overrides. Traditional approaches such as inverse optimization and constraint learning tend to overfit sparse data and produce complex formulations. Our central idea is to combine the reasoning capabilities and prior knowledge of LLMs with the formal grounding provided by optimization. We realize this idea through two complementary paradigms. Top-down, an LLM proposes structural repairs, including new constraints and variables, whose numerical parameters are calibrated and validated through optimization. Bottom-up, optimization infers cuts from observed decisions, which the LLM contextualizes into interpretable, generalizable modeling constraints. We evaluate our approach on 38 misspecification scenarios spanning nine classes of optimization problems, several drawn from real-world applications, and show that TACIT can repair 78.9% of them (vs. 60.5% for the best baseline).
Abstract:Real-world optimization problems are difficult to model accurately because many objectives and constraints reside in domain experts' tacit knowledge, making them hard to formalize. As a result, optimization models often contain miscalibrated objectives, missing constraints, or omitted decision variables, leading to solutions that fail to reflect operational realities. We address this challenge by automatically repairing misspecified formulations using historical data consisting of past solutions and subsequent user overrides. Traditional approaches such as inverse optimization and constraint learning tend to overfit sparse data and produce complex formulations. Our central idea is to combine the reasoning capabilities and prior knowledge of LLMs with the formal grounding provided by optimization. We realize this idea through two complementary paradigms. Top-down, an LLM proposes structural repairs, including new constraints and variables, whose numerical parameters are calibrated and validated through optimization. Bottom-up, optimization infers cuts from observed decisions, which the LLM contextualizes into interpretable, generalizable modeling constraints. We evaluate our approach on 38 misspecification scenarios spanning nine classes of optimization problems, several drawn from real-world applications, and show that TACIT can repair 78.9% of them (vs. 60.5% for the best baseline).
Optimization and Control (math.OC); Artificial Intelligence (cs.AI)
arXiv:2609.38434 [math.OC]
(or arXiv:2609.38434v1 [math.OC] for this version)
https://doi.org/10.48550/arXiv.2609.38434 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 283 words
Related coverage
Links #5: 2026/07
LessWrong · Jul 31, 2026
Game Design Creativity evals
LessWrong · Oct 12, 2026
The Alignment Problem: Computer Science Meets Analytic Philosophy
LessWrong · Oct 18, 2026
The art and science of hyperparameter optimization on Amazon Nova Forge
AWS · Jun 2, 2026