Speakers
Eric Anton Moreno
(Massachusetts Institute of Technology (US))
Nan Huang
(UC San Diego)
Yue Ma
Description
LLM agents are reshaping how particle physicists interact with data, code, and the research process itself. From "vibe coding" to autonomous scientist frameworks capable of hypothesis generation and experiment design, these tools are beginning to transform workflows in high-energy physics.
This BoF session invites practitioners, skeptics, and enthusiasts to share experiences and openly discuss the opportunities and challenges of deploying LLM agents in real HEP workflows. Key questions we hope to explore together include:
- What does vibe coding look like in practice for HEP analysis pipelines, and where does it break down?
- How close are we to genuinely autonomous scientific agents in particle physics, and what are the current bottlenecks?
- What are the risks around hallucinated results, reproducibility, and loss of domain expertise, and how do we mitigate them?
- What tooling, benchmarks, or community standards are needed to evaluate and trust agent-generated science?
A few slides will help guide the discussion, but the session is primarily driven by open conversation. Whether you have shipped an agent-assisted workflow or remain deeply skeptical, your perspective belongs in the room.
| Do you plan to submit a 4-page extended abstract on OpenReview (only for Presentations/Posters)? | No |
|---|