arXiv cs.AI· Antonio J. Costa, Caterina Doglioni, Christian G\"utschow, Andrew D. Pilkington, Sukanya Sinha·· 12 小时前AI 评分31
AgentRivet：用 LLM 自动从期刊论文生成 Rivet 分析例程的系统
AgentRivet: an automated system for producing Rivet routines from journal publications
AI 导读
AgentRivet 是一套基于大语言模型的多步自动化工作流，可从已发表论文中提取物理分析信息并生成缺失的 Rivet 例程，中间包含代码与物理双重自动审查。研究使用 OpenAI、Anthropic 和 Google 的商业 LLM 对 ATLAS 与 CMS 的两项近期测量进行测试，生成的例程语法错误很少、物理保真度合理，但仍有物理实现问题，多数源于论文中细微而含糊的定义。
正文 · 原文
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 allow 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.
| Subjects: | High Energy Physics - Experiment (hep-ex); Artificial Intelligence (cs.AI); High Energy Physics - Phenomenology (hep-ph) |
| Report number: | MCNET-26-14 |
| Cite as: | arXiv:2606.13535 [hep-ex] |
| (or arXiv:2606.13535v5 [hep-ex] for this version) | |
| https://doi.org/10.48550/arXiv.2606.13535 arXiv-issued DOI via DataCite |
Submission history
From: Christian Gütschow [view email]
[v1]
Thu, 11 Jun 2026 16:22:40 UTC (1,637 KB)
[v2]
Thu, 18 Jun 2026 19:48:22 UTC (1,639 KB)
[v3]
Tue, 23 Jun 2026 14:52:24 UTC (1,646 KB)
[v4]
Mon, 14 Sep 2026 17:08:11 UTC (2,007 KB)
[v5]
Fri, 16 Oct 2026 15:41:26 UTC (2,007 KB)
来源：arXiv cs.AI · arxiv.org