Reranking models designed for molecular information retrieval systems.
Derify
company
AI & ML interests
Cheminformatics | Chemical Language Models | Molecular Information Retrieval
Recent Activity
View all activity
SMILES Matryoshka Representation Learning Embedding Transformer
A set of SMILES datasets canonicalized with RDKit and 33% randomly augmented for robust, diverse molecular ML training.
Reranking models designed for molecular information retrieval systems.
ModernBERT as a Chemical Language Model
SMILES Matryoshka Representation Learning Embedding Transformer
A set of SMILES datasets canonicalized with RDKit and 33% randomly augmented for robust, diverse molecular ML training.
SAFE (Sequential Attachment-based Fragment Embedding) representation datasets with corresponding SMILES strings