Fast and scalable retrosynthetic planning with a transformer neural network and speculative beam search

Digital Discovery 2026 Open access

Read full text
Authors: Natalia G. Andronova, Mikhail Grigor'evich Andronov, Jürgen Schmidhuber, Michael D. Wand, Djork-Arné Clevert
DOI: 10.1039/d5dd00573f
Cited by: 1
References: 26

Abstract

Medusa model and speculative beam search accelerate template-free synthesis planning.

Fields of Study & Subjects

Machine Learning in Materials Science Block Copolymer Self-Assembly Neural Networks and Reservoir Computing

Publication details

Journal / source
Digital Discovery
Publisher
Royal Society of Chemistry (RSC)
Volume
5
Issue
4
Pages
1783–1793
Publication date
1 January 2026
Language
EN