BREAKING NEWS
The Royal Swedish Academy of Sciences has decided to award the 2026 #NobelPrize in Chemistry to Henri B. Kagan and Kenso Soai “for the discovery of non-linear effects and autocatalysis in asymmetric organic synthesis.”
My research, "Catalyst-Specialized Chemical Language Model Based on Transformer Variational Autoencoder for Catalyst Design and Discovery," has been published! Read it here: https://t.co/X8kbvLtfeB #PublishedArticle@wileyinresearch
@AdvPortfolio@ohuelab We combined #CatTransVAE, template-guided generation, property prediction, and Bayesian optimization to explore catalysts for Suzuki–Miyaura coupling. Novel candidates were generated and evaluated, providing a step toward AI-assisted catalyst discovery.
Our paper, #CatTransVAE, a catalyst-specialized chemical language model with template-guided generation framework for catalyst design and discovery, is now published in Advanced Intelligent Discovery! @AdvPortfolio@ohuelab ⚗️💡⚙️
📄 Read paper here: https://t.co/7D6shNrooP
@AdvPortfolio@ohuelab Catalysts occupy a distinct chemical space with complex structures, unique elemental compositions and coordination, and diverse molecular properties.
We developed #CatTransVAE, a new catalyst-specific model that learns a broader catalyst latent space.
Our new paper has been published in Advanced Intelligent Discovery!
In this work with Apakorn (@appaeak), we introduce CatTransVAE, a catalyst-specialized chemical language model for catalyst design and discovery.
The model combines pretraining on general compounds with fine-tuning on diverse catalyst datasets to better capture catalyst-specific chemical space. We also explore template-guided molecular generation, property prediction using learned embeddings, and a catalyst optimization workflow based on Bayesian optimization.
Catalyst-Specialized Chemical Language Model Based on Transformer Variational Autoencoder for Catalyst Design and Discovery
https://t.co/dGQLwaSPW7
#AI #MachineLearning #Chemoinformatics #CatalystDesign #GenerativeAI
It’s getting hard to keep track of all the new open-source cofolding models coming out. To help everyone stay up to date, I'm excited to share https://t.co/Gc9PaF6GoY, a PDB-synced leaderboard of open models, updated weekly. All predictions are available to view & download! (1/4)
🚀 Delighted to share our latest preprint on #CatTransVAE — Catalyst-specialized chemical language model based on transformer-VAE designed for catalyst generation and discovery. Now out on @ChemRxiv
📄: https://t.co/PT8uzXPXkT
#Catalyst#GenerativeAI#MaterialsInformatics
Catalyst-specialized Chemical Language Model Based on Transformer Variational Autoencoder for Catalyst Design and Discovery | ChemRxiv https://t.co/yo00kjNGZ1
久しぶりのChemRxivだよ w/@appaeak
New Anthropic Science Blog: Making Claude a chemist.
To manipulate a molecule, chemists first need to understand its structure. Their main tool is NMR spectroscopy.
We found Opus 4.7 matches—and on some tasks beats—dedicated NMR software. Read more: https://t.co/1jUvz7wdhV
🆕AI for Science萌芽的挑戦研究創出事業(SPReAD)、6月2日より第2回公募を開始予定!
人文学、社会科学を含むあらゆる分野において、AIを利活用した研究を加速することを目的に、全国で1,000件のAI for Scienceプロジェクトへのチャレンジを支援します。
▼公式サイト
https://t.co/XekLXo6h51