aiqbench analysis
Mol-JEPA advances molecular foundation models by addressing key limitations like modality collapse through a multimodal Joint Embedding Predictive Architecture. By integrating diverse data sources including quantum chemistry simulations and ADMET profiles, the framework creates robust latent representations for drug discovery tasks. This approach offers a scalable solution to improve biochemical context understanding beyond traditional perturbation-based methods.
Mol-JEPA: A multimodal Joint Embedding Predictive Architecture for Molecules
arXiv:2608.22642v1 Announce Type: cross 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.
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.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2608.22642 [cs.LG]
(or arXiv:2608.22642v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.22642 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 192 words
Related coverage
MDN-VJEPA: Multimodal Probabilistic Joint-Embedding World Models
arXiv · Oct 14, 2026
Akasha 2: Hamiltonian State Space Duality and Visual-Language Joint Embedding Predictive Architectur
arXiv · Jun 16, 2026
World Models and Yann LeCun’s Bet Against Large Language Models
Quantum Zeitgeist · Jul 17, 2026
StageWAM: Joint-Embedding Stage Prediction for World-Action Models in Robot Manipulation
arXiv · Aug 17, 2026