AgentRivet: An automated system for producing Rivet routines from journal publications
Open Access
- 20 November 2026
- journal article
- Published by Stichting SciPost in SciPost Physics Codebases
Abstract
Particle physics collider experiments provide Rivet routines as part of the analysis preservation strategy for model-independent measurements. Rivet is a C++ toolkit that allows new theoretical models to be compared to the measurements, thus aiding the development and tuning of Monte Carlo event generators as well as searches for physics beyond the Standard Model. However, analysis coverage is known to be incomplete, with only 39% of measurements having documented and publicly available Rivet routines. In this article, we design and implement an automated workflow based on Large Language Models with the goal of providing the missing routines. This multi-step workflow, referred to as AgentRivet, extracts the physics analysis information from published papers and writes the missing Rivet routines, with intermediate code- and physics- reviews as part of an autonomous quality control. We report the results obtained using commercial Large Language Models, provided by OpenAI, Anthropic, and Google, for two recent measurements from the ATLAS and CMS experiments. We find that AgentRivet produces competent Rivet routines with few syntax errors. The physics fidelity of the routines is reasonable and follows the explanations given in the relevant publications. Nevertheless, physics-implementation issues do arise and are investigated using the artefacts produced by AgentRivet. The majority of physics implementation issues arise from subtle-but-ambiguous definitions in the given publication, although some models struggle to implement complex observables even when clear definitions are given.Funding Information
- European Research Council (101002463)
- Science and Technology Facilities Council (UKRI:2850)
- Science and Technology Facilities Council (UKRI:3900)
This publication has 43 references indexed in Scilit:
- Measurement of event shape variables using charged particles inside jets in proton-proton collisions at $ \sqrt{s}=13 $ TeVJournal of High Energy Physics, 2026
- Precise measurement of the $t\bar{t}$ production cross-section and lepton differential distributions in $ e\mu $ dilepton events from $\sqrt{s}=13\,\text {TeV}$ pp collisions with the ATLAS detectorThe European Physical Journal C, 2026
- Agents of discoverySciPost Physics, 2026
- Robust independent validation of experiment and theory: Rivet version 4 release noteSciPost Physics Codebases, 2024
- Codebase release 4.0 for RIVETSciPost Physics Codebases, 2024
- A survey on large language model based autonomous agentsFrontiers of Computer Science, 2024
- A comprehensive guide to the physics and usage of PYTHIA 8.3SciPost Physics Codebases, 2022
- Reinterpretation of LHC Results for New Physics: Status and recommendations after Run 2SciPost Physics, 2020
- HEPData: a repository for high energy physics dataJournal of Physics: Conference Series, 2017
- The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulationsJournal of High Energy Physics, 2014