Selected publications — Gaia Grosso

Peer-reviewed journal articles

Learning to validate generative models: a goodness-of-fit approach

Cappelli P.^†, Grosso G., Letizia M., Reyes-Gonzalez H., Zanetti M.

Machine Learning: Science and Technology 7, 045011 (2026) · DOI · arXiv:2511.09118

Anomaly-preserving contrastive neural embeddings for end-to-end model-independent searches at the LHC

Metzger K.^†, Sodini M., Xu L.^†, Aarrestad T., Govorkova E., Grosso G., Harris P.

Physical Review D 112, 072011 (2025) · DOI

Robust resonant anomaly detection with NPLM

Grosso G., Sengupta D., Golling T., Harris P.

European Physical Journal C 85, 1074 (2025) · DOI

Multiple testing for signal-agnostic searches for new physics with machine learning

Grosso G.^*, Letizia M.^*

European Physical Journal C 85, 4 (2025) · DOI · arXiv:2408.12296

Anomaly-aware summary statistic from data batches

Grosso G. single author

Journal of High Energy Physics 12 (2024) 093 · DOI · arXiv:2407.01249

Goodness of fit by Neyman–Pearson testing

Grosso G., Letizia M., Wulzer A., Pierini M.

SciPost Physics 16, 123 (2024) · DOI

Fast kernel methods for data quality monitoring as a goodness-of-fit test

Grosso G., Lai N.^†, Letizia M., Pazzini J., Rando M., Rosasco L., Wulzer A., Zanetti M.^α

Machine Learning: Science and Technology 4, 035029 (2023) · DOI

Learning new physics efficiently with nonparametric methods

Letizia M., Losapio G., Rando M., Grosso G., Wulzer A., Pierini M., Zanetti M., Rosasco L.

European Physical Journal C 82, 879 (2022) · DOI

Learning new physics from an imperfect machine

D’Agnolo R.T., Grosso G., Pierini M., Wulzer A., Zanetti M.^α

European Physical Journal C 82, 275 (2022) · DOI · arXiv:2111.13633

Learning multivariate new physics

D’Agnolo R.T., Grosso G., Pierini M., Wulzer A., Zanetti M.^α

European Physical Journal C 81, 89 (2021) · DOI

Peer-reviewed conference papers and proceedings

AutoSciDACT: Automated scientific discovery through contrastive embedding and hypothesis testing

Bright-Thonney S., Reissel C., Grosso G., Woodward N.^†, Govorkova E., Novak A., Park S.E., Moreno E., Harris P.

Advances in Neural Information Processing Systems 38, 88575–88614 (NeurIPS 2025) · DOI · Proceedings PDF · arXiv:2510.21935

Triggerless data acquisition pipeline for machine learning based statistical anomaly detection

Grosso G., Lai N.^†, Migliorini M., Pazzini J., Triossi A., Zanetti M., Zucchetta A.^α

EPJ Web of Conferences 295, 02033 (2024) — Proceedings of CHEP 2023 · DOI

Preprints and papers under review

An end-to-end framework for uncertainty-aware validation of generative models

Badaracco G.^†, Benevedes S., Reissel C., Harris P., Grosso G.

Preprint, 2026 · arXiv:2610.37398

Covariant contrastive learning for uncertainty-aware anomaly detection

Tong S.^†, Harris P., Grosso G.

Preprint, 2026 · arXiv:2609.35755

Sparse, self-organizing ensembles of local kernels detect rare statistical anomalies

Grosso G., Hindupur S.S.R.^†, Fel T., Bright-Thonney S., Harris P., Ba D.

Preprint, 2025 · arXiv:2511.03095

Product manifold machine learning for physics

Woodward N.S.^†, Park S.E., Grosso G., Krupa J., Harris P.

Preprint, 2024 · arXiv:2412.07033

Reviews, perspectives and white papers

The look-elsewhere effect in searches for new physics: from classical methods to machine learning

Grosso G., Rodriguez-Benitez S.^†, Lyons L.

Review, in preparation

VERAIPHY — Validation & Evaluation for Robust AI in PHYsics

Grosso G., Winterhalder R., Brenner L., Lyons L., Plehn T.

Opening article of the PHYSTAT VERAIPHY series, 2026 · arXiv:2608.17724 · Series

Are we ready for AI-driven discovery? AI verification before the next fundamental physics breakthrough

Grosso G., Mikuni V., Heinrich L.

Review, VERAIPHY series, 2026 — to be submitted to SciPost Physics · arXiv:2607.10039

A virtuous cycle: generative AI and discovery in the physical sciences

Grosso G., Harris P., Mishra-Sharma S., Shanahan P.

An MIT Exploration of Generative AI (2024) · DOI

Workshop papers (non-archival)

Adaptive and interpretable two-sample test as a batch-level tool for AI safety

Grosso G., Marin-Llobet A., Ba D.

ICML 2026 Workshop on Mechanistic Interpretability, Seoul · Paper

An end-to-end pipeline for uncertainty-aware validation of generative AI

Badaracco G.^†, Benevedes S., Grosso G., Reissel C., Aarrestad T., Harris P.

NeurIPS 2025 Workshop on Machine Learning and the Physical Sciences, San Diego · Paper

Learning dictionaries of new physics with sparse local kernels

Grosso G., Govorkova E., Harris P., Moreno E., Raikman R.

NeurIPS 2024 Workshop on Machine Learning and the Physical Sciences, Vancouver · Paper

Collaboration papers

Member of the CMS Collaboration since 2018; on the author list since 2021 and co-author of 465 CMS Collaboration papers. Full list on INSPIRE-HEP.

Patent

Methods for explainable novelty discovery using artificial intelligence

Grosso G., Ba D.E., Hindupur Ravindra S.S.^†

U.S. Provisional Patent Application No. 63/909,442, filed October 31, 2025. Covers the SparKer method.