Configurable
Describe samples, parameters, likelihood choices, and minimizer behavior in structured input files.
GUNDAM · High Energy Physics
Generalized and Unified Neutrino Data Analysis Methods
A flexible framework for neutrino data analysis.
GUNDAM is a suite of configurable applications for statistical analysis. It brings reusable fitting, uncertainty, and validation tools together so improvements made for one analysis can benefit many.
Log-likelihood surfaceminimize −2 log(L)
Overview
One framework, many analyses
GUNDAM provides a common, configurable foundation for likelihood-based analyses. YAML and JSON inputs describe the job, allowing users to adapt an analysis without rewriting the framework for every study.
YAML / JSONConfiguration-driven workflows
C++17Preferred compiler standard
ROOT 6Core analysis ecosystem
Why GUNDAM?
Reusable analysis infrastructure
Analysis groups repeatedly solve the same hard problems: configuring models, propagating uncertainties, minimizing statistical objectives, validating fits, and presenting diagnostics. GUNDAM centralizes those tasks so improvements and fixes can be reviewed once and reused across analyses.
Extensible
Develop new analysis components on a shared foundation instead of starting from an isolated fitter.
Traceable
Keep analysis configuration alongside versioned source and input material for reproducible studies.
Collaborative
Let debugging and optimization performed for one use case benefit the broader user community.
Lineage and Legacy
Built from neutrino-analysis experience
GUNDAM began as a fork of xsllhFitter, developed by the T2K cross-section working group. It was initially developed for T2K near-detector fits and studies of the upgraded ND280 detectors, then expanded for a broader range of fitting and cross-section analyses.
Its lineage also reflects experience from the T2K BANFF near-detector beam and neutrino-flux fitter. The present project carries that experience forward in a more general, configurable framework.
Experiments
Growing across experimental programs
Developed in the T2K ecosystem, GUNDAM is now used by ICARUS and is actively promoted within DUNE.
Publications
Research using GUNDAM
Publicly available papers that use GUNDAM directly or document its statistical methods.
Measurement of muon neutrino charged-current quasielasticlike cross section using off-axis NuMI beam at ICARUS
The first ICARUS neutrino cross-section result uses GUNDAM for fitting and cross-section extraction.
Open publicationFlow-Based Surrogates for High-Dimensional Likelihoods in Experimental Neutrino Physics
Develops portable likelihood surrogates and an interface to GUNDAM for high-dimensional near-detector fits.
Open publicationPosters
Conference and workshop posters
Posters focused on GUNDAM or analyses in which it is a central fitting framework.
GUNDAM: a pioneering universal tool for long-baseline neutrino oscillation experiments
An overview of GUNDAM’s architecture, performance, analysis workflow, and T2K applications.
Open posterGUNDAM: a next-generation fitting tool for future T2K analyses
Introduces the framework, configuration model, performance, and early T2K analysis examples.
Open posterTalks
Presentations and training material
Public presentation material on GUNDAM and analyses that use it.
Results from the T2K Near Detector Analysis Fit and Prospects from Near Detector Upgrade Data
Reviews the GUNDAM and MaCh3 cross-validation of the latest T2K near-detector constraints.
Open slidesICARUS cross sections using the off-axis NuMI beam at Fermilab
Describes the first published ICARUS physics analysis completed with GUNDAM.
Open slidesGUNDAM fitter for T2K
A dedicated introduction to the fitter and its role in T2K statistical analyses.
Open talk materialsLicense
Open scientific software
GUNDAM source files without a more specific license are distributed under the GNU Lesser General Public License, version 2.1. Consult the upstream license file before redistributing or modifying the software.