Paper Image

Language models for catalyst discovery

Published on:

22 October 2023

Primary Category:

Artificial Intelligence

Paper Authors:

Henry W. Sprueill,

Carl Edwards,

Mariefel V. Olarte,

Udishnu Sanyal,

Heng Ji,

Sutanay Choudhury


Key Details

Proposes querying large language models to suggest catalyst candidates

Models prompted iteratively with constraints to refine suggestions

Introduces two new benchmark datasets for evaluating model reasoning

Shows improved performance over chain-of-thought prompting baselines

Demonstrates models can produce scientifically valid explanations

AI generated summary

Language models for catalyst discovery

This paper proposes using large language models to suggest novel catalysts for chemical reactions. The models are queried in a stepwise fashion to narrow down promising candidates based on key properties like cost and stability. This approach augments human reasoning for accelerated scientific discovery.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up