AgentRivet: an automated system for producing Rivet routines from journal publications
作者: Antonio J. Costa, Caterina Doglioni, Christian Gütschow, Andrew D. Pilkington, Sukanya Sinha
分类: cs.AI
发布日期: 2026-06-12
💡 一句话要点
提出AgentRivet以自动生成Rivet例程解决分析覆盖不足问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 粒子物理 Rivet例程 自动化系统 大型语言模型 分析覆盖 物理分析
📋 核心要点
- 现有的Rivet例程覆盖不足,只有39%的测量结果有文档化的例程,影响了分析的完整性。
- 论文提出的AgentRivet利用大型语言模型自动提取物理分析信息并生成Rivet例程,旨在填补这一空白。
- 实验结果显示,AgentRivet生成的Rivet例程语法错误较少,物理保真度合理,但仍需解决一些物理实现问题。
📝 摘要(中文)
粒子物理对撞机实验提供Rivet例程作为模型无关测量的分析保存策略。然而,目前仅有39%的测量结果有文档化的Rivet例程。本文设计并实现了一个基于大型语言模型的自动化工作流程AgentRivet,旨在提供缺失的Rivet例程。该多步骤工作流程从已发表论文中提取物理分析信息,并编写缺失的Rivet例程,同时进行代码和物理审查以确保质量。实验结果表明,AgentRivet生成的Rivet例程具有较少的语法错误,物理保真度合理,但仍存在物理实现问题,主要源于文献中的模糊定义。
🔬 方法详解
问题定义:本文旨在解决粒子物理实验中Rivet例程的缺失问题,现有方法的痛点在于只有39%的测量结果有文档化的例程,导致分析覆盖不足。
核心思路:AgentRivet的核心思路是利用大型语言模型自动提取已发表论文中的物理分析信息,并生成相应的Rivet例程,从而提高分析的完整性和效率。
技术框架:AgentRivet的整体架构包括多个模块,首先是信息提取模块,从文献中提取相关的物理分析信息;其次是例程生成模块,根据提取的信息生成Rivet例程;最后是质量控制模块,进行代码和物理审查,确保生成例程的质量。
关键创新:该研究的关键创新在于将大型语言模型应用于物理分析例程的自动生成,显著提高了生成效率和准确性,与传统手动编写方法相比,减少了人力成本和时间消耗。
关键设计:在设计中,AgentRivet使用了商业大型语言模型,如OpenAI、Anthropic和Google提供的模型,确保了生成内容的质量和准确性。同时,设置了多层次的审查机制,以识别和修正生成过程中的潜在错误。
🖼️ 关键图片
📊 实验亮点
实验结果表明,AgentRivet生成的Rivet例程在语法错误方面表现良好,错误率较低。物理保真度合理,符合相关文献的解释,但仍需解决一些由模糊定义引起的物理实现问题。
🎯 应用场景
AgentRivet的研究成果在粒子物理领域具有广泛的应用潜力,能够为实验分析提供自动化支持,提升分析的效率和准确性。未来,该系统可以扩展到其他领域的自动化分析任务,推动科学研究的进展。
📄 摘要(原文)
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.