Mol-JEPA: A multimodal Joint Embedding Predictive Architecture for Molecules

2026-08-23

AI Summary

本文提出Mol-JEPA框架，通过模态掩码利用多种药物发现数据解决分子基础模型中的化学无效增强、模态崩溃等问题。

Abstract

Despite recent advances in molecular foundation models, several limitations remain, such as chemically invalid augmentations, modality collapse, and incomplete representation of biochemical environments. To address these challenges, we present \textbf{Mol-JEPA}, a scalable framework for learning molecular world models. Rather than relying on suboptimal molecular perturbations, our model uses modality masking to exploit information from molecular structures, cellular phenotypes, binding affinities, ADMET profiles, quantum chemistry simulations and other drug discovery data. Across various benchmarks, we show that the representations learned by Mol-JEPA deliver strong performance, demonstrating the value of incorporating biochemical context through latent space prediction.

Problem

• chemically invalid augmentations
• modality collapse
• incomplete representation of biochemical environments

Innovation

• multimodal joint embedding
• modality masking
• molecular world models
• latent space prediction
• biochemical context