arXiv Machine Learning September 1, 2026 By David Sulu, Lorenzo Di Fruscia, Jana M. Weber

Structural Hierarchy and Geometry in Molecular Representation Learning

Read the original on arXiv Machine Learning →

The paper investigates how explicitly supervising molecular embeddings with a molecule’s Bemis‑Murcko scaffold influences representation learning. Experiments compare Euclidean and Lorentz contrastive objectives under two augmentation strengths, showing that scaffold‑supervised models consistently group molecules by identical and related scaffolds. These embeddings also enhance property prediction on several tasks, though the magnitude of improvement varies with the target property and the geometry used.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Jul 3

Probing Chemical Language Models: Effects of Pre-training and Fine-tuning

arXiv:2607. 02140v1 Announce Type: new Abstract: Chemical language models (CLMs) are trained with linearized representations such as SMILES, yet it remains unclear which chemically meaningful substructures they encode.

By Anna Karnysheva, Dietrich Klakow, Ji-Ung Lee

llms fine-tuning

arXiv AI

Aug 19

Domain-Adapted Molecular Language Models for Efficient Search of Make-on-Demand Libraries

The study evaluates four pretrained molecular language models on six virtual libraries covering drug discovery, organic materials, and catalysis. It finds that native embeddings vary widely in performance, while molecular fingerprints remain consistently strong. Fine‑tuning the models on library‑specific data markedly improves sample efficiency, with several adapted encoders outperforming others across all tasks.

By Henrik Wille, Luis-Finley Sch\"utz, Felix Strieth-Kalthoff

llms rag robotics fine-tuning benchmarks

arXiv AI

Aug 25

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

Mol-JEPA is a scalable multimodal framework that learns molecular world models by using modality masking instead of suboptimal perturbations. It incorporates diverse data such as molecular structures, cellular phenotypes, binding affinities, ADMET profiles, quantum chemistry simulations, and other drug‑discovery information. Benchmarks show that the representations it learns perform strongly, highlighting the benefit of embedding biochemical context via latent‑space prediction.

By Florian Rottach, Sebastian Schieferdecker, William Rudman, Randall Balestriero, Carsten Eickhoff

rag diffusion multimodal benchmarks

arXiv AI

Aug 5

Learning Molecular Representations from Cellular Phenotypes with Structure Preservation

arXiv:2608. 02688v1 Announce Type: cross Abstract: Phenotypic drug discovery enables the discovery of functional relationships between molecular structures and cellular responses.

By Xuan Lin, Jingyu Sheng, Tengfei Ma, Li Sun, Dapeng Xiong

rag multimodal safety

Hugging Face Trending Papers

Aug 11

Multi-Granular Rationale-Guided Molecular LLM for Property Prediction

Large language models (LLMs) are widely applied across chemical tasks, such as molecular property prediction, which underpins drug discovery. Molecular LLMs represent a molecule through several modalities, notably a 1D SMILES sequence or a 2D molecular graph.

llms fine-tuning

arXiv Machine Learning

Jul 3

Do LLMs Truly Generalize in the Molecular Domain? A Perturbation-Based Analysis

arXiv:2607. 01800v1 Announce Type: new Abstract: Large Language Models (LLMs) have recently shown promise in molecular discovery, yet a gap remains between their probabilistic nature over discrete sequential tokens and the rigid topological constraints of chemical space.

By Jiatong Li, Weida Wang, Changmeng Zheng, Shufei Zhang, Yatao Bian, Xiao-yong Wei, Qing Li

llms

One email a morning, machine-written

Related stories

Probing Chemical Language Models: Effects of Pre-training and Fine-tuning

Domain-Adapted Molecular Language Models for Efficient Search of Make-on-Demand Libraries

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

Learning Molecular Representations from Cellular Phenotypes with Structure Preservation

Multi-Granular Rationale-Guided Molecular LLM for Property Prediction

Do LLMs Truly Generalize in the Molecular Domain? A Perturbation-Based Analysis