SciPost Submission Page
Iterative HOMER with uncertainties
by Anja Butter, Ayodele Ore, Sofia Palacios Schweitzer, Tilman Plehn, Benoît Assi, Christian Bierlich, Philip Ilten, Tony Menzo, Stephen Mrenna, Manuel Szewc, Michael K. Wilkinson, Ahmed Youssef, Jure Zupan
Submission summary
| Authors (as registered SciPost users): | Ayodele Ore · Sofia Palacios Schweitzer · Tilman Plehn · Manuel Szewc |
| Submission information | |
|---|---|
| Preprint Link: | https://arxiv.org/abs/2509.03592v1 (pdf) |
| Code repository: | https://github.com/ayo-ore/iterative-homer |
| Date submitted: | Sept. 16, 2025, 5:29 p.m. |
| Submitted by: | Ayodele Ore |
| Submitted to: | SciPost Physics |
| Ontological classification | |
|---|---|
| Academic field: | Physics |
| Specialties: |
|
| Approaches: | Computational, Phenomenological |
Abstract
We present iHOMER, an iterative version of the HOMER method to extract Lund fragmentation functions from experimental data. Through iterations, we address the information gap between latent and observable phase spaces and systematically remove bias. To quantify uncertainties on the inferred weights, we use a combination of Bayesian neural networks and uncertainty-aware regression. We find that the combination of iterations and uncertainty quantification produces well-calibrated weights that accurately reproduce the data distribution. A parametric closure test shows that the iteratively learned fragmentation function is compatible with the true fragmentation function.
Author indications on fulfilling journal expectations
- Provide a novel and synergetic link between different research areas.
- Open a new pathway in an existing or a new research direction, with clear potential for multi-pronged follow-up work
- Detail a groundbreaking theoretical/experimental/computational discovery
- Present a breakthrough on a previously-identified and long-standing research stumbling block
Current status:
Reports on this Submission
Strengths
- Clear description of the new method.
- Proper treatment of uncertainties.
- The method is physically interpretable.
- Rigorous validation through closure test example.
- Publicly accessible code for reproducibility.
Weaknesses
- Dependence on a specific hadronization model (string model).
- Lack of application to real data (or mismodeled data).
- No benchmark on computing time shown.
Report
This is a solid and technically sound piece of work. I recommend publication, provided that the authors address a few minor comments.
Requested changes
-
Do I understand correctly that the method is specific to the string model? If so, could you comment on the limitation on the flexibility coming from the string model? And would it be possible to adapt the method to other hadronization model, e.g. cluster?
-
The method assumes negligible systematic uncertainty in the reweighing step 1 and negligible statistical uncertainty in the factorization step 2. Is this always guaranteed? What would one need to do if otherwise?
-
Could you explain further how using the Machine Learning based hadronization models could improve MCEG accuracy?
-
Could you provide any benchmarks on the training / computing time with the proposed method?
Recommendation
Ask for minor revision
Strengths
2 - Contains all necessary details and a detailed demonstration of the performance.
3 - Has the potential for other applications in the field.
Weaknesses
Report
Iterative HOMER with uncertainties
by Anja Butter et al.
A majority of experimental analyses of collider data are currently
based on a comparison with Monte Carlo simulations for which
the LUND hadronization model are used. This requires fragmentation
functions as input which have to be determined from data. The
authors present a method to extract fragmentation functions using
a combination of Bayesian neural networks and regression. The
submitted paper is the third in a series and presents two important
improvements over the previous versions: an iterative procedure
and modifications in the architectural choices lead to a better
assessment of both accuracy and precision for the extracted
fragmentation function.
The paper is well written and contains all necessary details, as well
as a detailed demonstration of the performance achieved. I expect
that the information given in the paper together with the code,
which is made available on github.com, will allow other workers
in the field to benefit also for other applications.
I support publication of the submitted article without modifications.
Requested changes
None
Recommendation
Publish (surpasses expectations and criteria for this Journal; among top 10%)