ChemBERTa-3: an open source training framework for chemical foundation models
Authors:
Top Cited Papers
Open Access
- 19 January 2026
- journal article
- research article
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (2) , 662-685
Abstract
An overview of ChemBERTa-3, an open-source training and benchmarking framework designed to train and fine-tune large-scale chemical foundation models.Keywords
All Related Versions
Funding Information
- U.S. Department of Energy (DE-AC52-07NA27344)
- Defense Threat Reduction Agency (HDTRA13081-40035)
This publication has 20 references indexed in Scilit:
- A Comparative Evaluation of Machine Learning and Deep Graph Learning for Chemical Ecotoxicological PredictionACS Omega, 2025
- Transformers and Large Language Models for Chemistry and Drug Discovery2024
- Chemprop: A Machine Learning Package for Chemical Property PredictionJournal of Chemical Information and Modeling, 2023
- Advanced deep learning methods for molecular property predictionQuantitative Biology, 2023
- Exposing the Limitations of Molecular Machine Learning with Activity CliffsJournal of Chemical Information and Modeling, 2022
- ZINC20—A Free Ultralarge-Scale Chemical Database for Ligand DiscoveryJournal of Chemical Information and Modeling, 2020
- AMPL: A Data-Driven Modeling Pipeline for Drug DiscoveryJournal of Chemical Information and Modeling, 2020
- Improving Conformer Generation for Small Rings and Macrocycles Based on Distance Geometry and Experimental Torsional-Angle PreferencesJournal of Chemical Information and Modeling, 2020
- Molecular property prediction: recent trends in the era of artificial intelligenceDrug Discovery Today: Technologies, 2019
- Analyzing Learned Molecular Representations for Property PredictionJournal of Chemical Information and Modeling, 2019