Speaker
Description
I present the current status of the NNFF2.0 project, aimed at a next-generation determination of fragmentation functions within the NNPDF framework. The analysis incorporates recent NNLO calculations for inclusive hadron production in deep-inelastic scattering and hadron–hadron collisions, and exploits exact NNLO theory predictions rather than K-factor approximations. On the methodological side, the fit is performed using an optimized neural-network architecture determined through a systematic hyperparameter scan. I discuss the associated software developments, including an improved interpolation-grid infrastructure and a new tool for the efficient computation of theory predictions required in the fit. Using existing FF sets as benchmarks, I assess the phenomenological impact of these theoretical and methodological improvements and outline their implications for the forthcoming NNFF2.0 release.
| Speaker confirmation | Yes |
|---|