Code and Data for Empirical and Synthetic Experiments on Formula L Optimization

This dataset provides the foundational resources for evaluating and optimizing Formula L , a novel mathematical framework for semantic-driven task allocation in multi-agent systems (MAS) powered by large language models (LLM). The dataset includes Python code and both empirical and synthetic data, specifically designed to validate the effectiveness of Formula L in improving task distribution, contextual relevance, and dynamic adaptation within MAS.The dataset comprises:Empirical data derived from real-world task allocation scenarios, demonstrating the practical application of Formula L  in domains such as logistics and autonomous systems.Synthetic data generated to simulate diverse and complex scenarios for benchmarking the scalability and robustness of the proposed framework.Python code implementing Formula L , along with detailed experiments and evaluation scripts.JSON files containing structured task definitions, historical context, and relevance metrics for both empirical and synthetic cases.This dataset supports reproducibility, further development, and comparative analysis for researchers and practitioners exploring task allocation strategies in MAS-LLM environments. By integrating semantic evaluation and optimization principles, it offers a robust foundation for advancing autonomous decision-making systems.