![[2509.08535] Agents of Discovery](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2509.08535] Agents of Discovery
Summary
The paper 'Agents of Discovery' explores the use of large language models (LLMs) as agents to automate data analysis in high energy physics, achieving performance comparable to human researchers.
Why It Matters
As data complexity in high energy physics increases, leveraging LLMs for automated analysis could significantly enhance research efficiency. This study demonstrates the potential of LLMs to perform complex tasks traditionally handled by human researchers, paving the way for future advancements in both AI and physics.
Key Takeaways
- LLMs can be utilized as agents to automate complex data analysis tasks.
- The study focuses on anomaly detection using the LHC Olympics dataset.
- Agent-created solutions achieved performance on par with state-of-the-art human results.
- This approach could alleviate the challenges posed by increasing data complexity in physics.
- The research highlights the evolving role of AI in scientific research.
High Energy Physics - Phenomenology arXiv:2509.08535 (hep-ph) [Submitted on 10 Sep 2025 (v1), last revised 17 Feb 2026 (this version, v2)] Title:Agents of Discovery Authors:Sascha Diefenbacher, Anna Hallin, Gregor Kasieczka, Michael Krämer, Anne Lauscher, Tim Lukas View a PDF of the paper titled Agents of Discovery, by Sascha Diefenbacher and 5 other authors View PDF HTML (experimental) Abstract:The substantial data volumes encountered in modern particle physics and other domains of fundamental physics research allow (and require) the use of increasingly complex data analysis tools and workflows. While the use of machine learning (ML) tools for data analysis has recently proliferated, these tools are typically special-purpose algorithms that rely, for example, on encoded physics knowledge to reach optimal performance. In this work, we investigate a new and orthogonal direction: Using recent progress in large language models (LLMs) to create a team of agents -- instances of LLMs with specific subtasks -- that jointly solve data analysis-based research problems in a way similar to how a human researcher might: by creating code to operate standard tools and libraries (including ML systems) and by building on results of previous iterations. If successful, such agent-based systems could be deployed to automate routine analysis components to counteract the increasing complexity of modern tool chains. To investigate the capabilities of current-generation commercial LLMs, we consi...
