This document presents the Neural Field Scattering Transform (NFST), a novel method for analyzing weak lensing convergence maps to constrain cosmological parameters. The NFST builds upon the Wavelet Scattering Transform (WST) by incorporating learnable neural field filters while maintaining rotational and translational symmetries, providing a balance between flexibility and robustness that is particularly effective with limited training data. The study demonstrates that the NFST outperforms the WST in predicting parameters like σ8 and w and in modeling posterior probability density when applied to simulated cosmological data. The researchers also introduce a new visualization technique to interpret the learned filters and coefficients, revealing that the NFST adapts to capture cosmologically relevant information, particularly in the non-Gaussian features of the large-scale structure.