Data from Discrete Flow-Based Generative Models for Measurement Optimization in Quantum Computing

Supporting dataset and models generated for "Discrete Flow-Based Generative Models for Measurement Optimization in Quantum Computing" arXiv:2509.15486 The data contains all sampled non-overlapping groupings generated with GFlowNets under full commutativity using the GINEw model, along with the GFN-ICS output files used to generate the results in Table 1 of "Discrete Flow-Based Generative Models for Measurement Optimization in Quantum Computing," in the folder Standard_FC_and_ICS.In the folder Composite_2qubit, we include all sampled graphs, outputs, and models for the different rewards listed in Table 5 of the main paper, which account for the number of measurements and two-qubit gates after compilation using the Markov-Hayes-Patel algorithm with all-to-all connectivity. The resulting models depend on the reward function, so the folders are organized accordingly for each molecule. For the supporting code to generate the data and perform the analysis from the paper, please refer to: https://github.com/ChemAI-Lab/GFlowNets-MOpt