Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2509.26201 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911400882864128 |
|---|---|
| author | Werbrouck, Andreas Lindsay, Marshall B. Maschmann, Matthew Young, Matthias J. |
| author_facet | Werbrouck, Andreas Lindsay, Marshall B. Maschmann, Matthew Young, Matthias J. |
| contents | Large Language Models (LLMs) have garnered significant attention for several years now. Recently, their use as independently reasoning agents has been proposed. In this work, we test the potential of such agents for knowledge discovery in materials science. We repurpose LangGraph's tool functionality to supply agents with a black box function to interrogate. In contrast to process optimization or performing specific, user-defined tasks, knowledge discovery consists of freely exploring the system, posing and verifying statements about the behavior of this black box, with the sole objective of generating and verifying generalizable statements. We provide proof of concept for this approach through a children's parlor game, demonstrating the role of trial-and-error and persistence in knowledge discovery, and the strong path-dependence of results. We then apply the same strategy to show that LLM agents can explore, discover, and exploit diverse chemical interactions in an advanced Atomic Layer Processing reactor simulation using intentionally limited probe capabilities without explicit instructions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_26201 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LLM Agents for Knowledge Discovery in Atomic Layer Processing Werbrouck, Andreas Lindsay, Marshall B. Maschmann, Matthew Young, Matthias J. Artificial Intelligence Mesoscale and Nanoscale Physics Materials Science Large Language Models (LLMs) have garnered significant attention for several years now. Recently, their use as independently reasoning agents has been proposed. In this work, we test the potential of such agents for knowledge discovery in materials science. We repurpose LangGraph's tool functionality to supply agents with a black box function to interrogate. In contrast to process optimization or performing specific, user-defined tasks, knowledge discovery consists of freely exploring the system, posing and verifying statements about the behavior of this black box, with the sole objective of generating and verifying generalizable statements. We provide proof of concept for this approach through a children's parlor game, demonstrating the role of trial-and-error and persistence in knowledge discovery, and the strong path-dependence of results. We then apply the same strategy to show that LLM agents can explore, discover, and exploit diverse chemical interactions in an advanced Atomic Layer Processing reactor simulation using intentionally limited probe capabilities without explicit instructions. |
| title | LLM Agents for Knowledge Discovery in Atomic Layer Processing |
| topic | Artificial Intelligence Mesoscale and Nanoscale Physics Materials Science |
| url | https://arxiv.org/abs/2509.26201 |