EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automating Chemical Reasoning in High‐Throughput Phase Identification With a Probabilistic, LLM‐Guided Framework

Olympia Dartsi, Lauren N Walters, Amalie E Trewartha, Steven B Torrisi, Amanda Volk, Liam Joyce, Gerbrand Ceder, Anubhav Jain

Advanced Science · 03.08.2026

DOI: 10.1002/advs.76450 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advanced Science, online erschienen am 03.08.2026. Verfasst von Olympia Dartsi und 7 weiteren Autor:innen. Beteiligte Einrichtung: Energy Technologies Area, Lawrence Berkeley National Laboratory, Berkeley, USA. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Toyota Research Institute.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Autonomous laboratories increasingly enable materials synthesis at scale, but traditional high‐throughput characterization workflows remain limited by the need for expert chemical intuition to distinguish plausible interpretations from formally good but chemically incorrect fits. We present an automated interpretation framework that combines probabilistic inference with automated chemical reasoning for phase identification from powder x‐ray diffraction (PXRD). The framework evaluates multiple candidate interpretations using diffraction pattern‐based metrics. It then refines these likelihoods using chemically‐informed priors derived from composition balance and a large language model (LLM)–based plausibility estimate with human‐readable justification and also produces a trustworthiness score. In a blinded multi‐project benchmark, the framework's top‐ranked interpretation was selected over the lowest‐ baseline in 93% of cases where evaluators expressed a clear preference (95% CI: [78%, 98%], ). Trust decisions made by the framework aligned with expert judgment in approximately 75%–80% of cases. In a second evaluation, the framework systematically identified cases where lowest‐ interpretations were chemically implausible and surfaced credible alternatives to historically ambiguous samples. By reframing phase identification as a problem of probabilistic reasoning and trust‐aware decision making, this work demonstrates how chemical intuition can be automated and scaled.

Autor:innen und Institutionen

  • Energy Technologies Area, Lawrence Berkeley National Laboratory, Berkeley, USA.
  • Bakar Institute of Digital Materials for the Planet, University of California, Berkeley, USA.
  • Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, USA.
  • Energy & Materials Division, Toyota Research Institute, Los Altos, USA.
  • Department of Materials Science & Engineering, University of California, Berkeley, USA.

Förderung

  • Toyota Research Institute

Zitieren

APA

Dartsi, O., Walters, L. N., Trewartha, A. E., Torrisi, S. B., Volk, A., Joyce, L., Ceder, G., & Jain, A. (2026). Automating Chemical Reasoning in High‐Throughput Phase Identification With a Probabilistic, LLM‐Guided Framework. Advanced Science. https://doi.org/10.1002/advs.76450

Vancouver

Dartsi O, Walters LN, Trewartha AE, Torrisi SB, Volk A, Joyce L, et al. Automating Chemical Reasoning in High‐Throughput Phase Identification With a Probabilistic, LLM‐Guided Framework. Advanced Science. 2026. doi:10.1002/advs.76450

BibTeX
@article{Dartsi2026,
  author = {Dartsi, Olympia and Walters, Lauren N and Trewartha, Amalie E and Torrisi, Steven B and Volk, Amanda and Joyce, Liam and Ceder, Gerbrand and Jain, Anubhav},
  title = {Automating Chemical Reasoning in High‐Throughput Phase Identification With a Probabilistic, LLM‐Guided Framework},
  journal = {Advanced Science},
  year = {2026},
  doi = {10.1002/advs.76450}
}
Kostenlos starten →