MolTRES: Improving Chemical Language Representation Learning for Molecular Property Prediction [EMNLP 2024]
Jun-Hyung Park, Yeachan Kim, Mingyu Lee, Hyuntae Park, and SangKeun Lee
Introduce new techniques based on generator-discriminator training and latent knowledge embedding to improve pre-trained chemical language models [paper]
This implementation is based on [IBM/MoLFormer].
- Python 3.10
- Pytorch 2.3.1
- Pytorch-lightning 2.2.4
- Pytorch-fast-transformers 0.4.0
- RDKit 2023.9.6