TRIDENT: Tri-Modal Molecular Representation Learning with Taxonomic Annotations and Local Correspondence

Feng Jiang (University of Texas at Arlington) · Mangal Prakash (Exscientia) · Hehuan Ma (University of Texas at Arlington) · Jianyuan Deng (Johnson and Johnson) · Yuzhi Guo (University of Texas at Arlington) · Maolaaisha Aminanmu (University of Zurich) · Tommaso Mansi (Johnson & Johnson Innovative Medicine) · Rui Liao (JnJ) · Junzhou Huang (University of Texas at Arlington)

functional semanticsgeometry-aware alignmentlocal alignment objectivemolecular property prediction molecular representationsmolecular substructuresmomentum-based mechanismmultimodal learning smilesstructure-function mappingssub-textual descriptionstaxonomic functional annotationstextual descriptionstri-modal featuresvolume-based alignment

Molecular property prediction aims to learn representations that map chemical structures to functional properties. While multimodal learning has emerged as a powerful paradigm to learn molecular representations, prior works have largely overlooked textual and taxonomic information of molecules for representation learning. We introduce TRIDENT, a novel framework that integrates molecular SMILES, textual descriptions, and taxonomic functional annotations to learn rich molecular representations. To achieve this, we curate a comprehensive dataset of molecule-text pairs with structured, multi-level functional annotations. Instead of relying on conventional contrastive loss, TRIDENT employs a volume-based alignment objective to jointly align tri-modal features at the global level, enabling soft, geometry-aware alignment across modalities. Additionally, TRIDENT introduces a novel local alignment objective that captures detailed relationships between molecular substructures and their corresponding sub-textual descriptions. A momentum-based mechanism dynamically balances global and local alignment, enabling the model to learn both broad functional semantics and fine-grained structure-function mappings. TRIDENT achieves state-of-the-art performance on 18 downstream tasks, demonstrating the value of combining SMILES, textual, and taxonomic functional annotations for molecular property prediction. Our code and data are available at https://github.com/uta-smile/TRIDENT.