Metraipolis › archive › dc_51568c253ccf

Can LLM Agents Improve Imperfect Earth System Model Parameterizations? — arXiv:2611.03543

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03543
License (as archived)
CC0-1.0
Archived
2026-11-03T11:26:51.105Z
SHA-256 of the archived text
bdce3ed3b6946b55ca0e5fc7ca0462ae8c1608721dd9a86c9f5fd686b85dc745
Stable URL of this page
https://metraipolis.com/documento/dc_51568c253ccf

How to cite

Can LLM Agents Improve Imperfect Earth System Model Parameterizations? — arXiv:2611.03543 — arXiv (en), https://arxiv.org/abs/2611.03543 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_51568c253ccf (Metraipolis archive dc_51568c253ccf, sha256 bdce3ed3b6946b55ca0e5fc7ca0462ae8c1608721dd9a86c9f5fd686b85dc745)
BibTeX
@misc{dc51568c253ccf,
  title = {Can LLM Agents Improve Imperfect Earth System Model Parameterizations? — arXiv:2611.03543},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03543},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_51568c253ccf; sha256 bdce3ed3b6946b55ca0e5fc7ca0462ae8c1608721dd9a86c9f5fd686b85dc745},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03543. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Earth system models (ESMs) rely on parameterizations to represent unresolved processes such as convection, cloud microphysics, and turbulence. However, translating advances in physical understanding into stable, validated parameterizations and integrating them into ESMs often takes years. Agentic AI could accelerate this process and enable more systematic exploration of design choices. Using warm-rain collision-coalescence as a test case, we evaluate agents' ability to synthesize scientific literature, reimplement published bulk parameterizations, and propose new functional forms as differentiable JAX code. We compare conventional parameterizations, data-driven emulators, and agent-generated closures against high-fidelity superdroplet simulations, repeating unattended experiments across ten language models. With expert guidance or iterative feedback, agents produced executable, differentiable, mass-conserving code, although unattended performance depended strongly on the scheme and language model. Iterative feedback lowered median rollout error for eight of ten models. Retuned physical schemes matched or outperformed many learned parameterizations at substantially lower computational cost, but two of 100 unattended agent-generated closures achieved slightly lower rain-mass error than the most accurate retuned scheme on held-out simulations.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Atmospheric and Oceanic Physics (physics.ao-ph) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.03543 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T11:02:37.918Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).