The paper reports a large-scale, compute-controlled study of Chemical Language Models (CLMs) involving over 30,000 experiments across different molecular representations, tokenizations, model sizes, datasets, and architectures. It finds clear scaling trends in pretraining loss but shows that these improvements do not translate into proportional gains in goal-directed molecular design, with chemical syntax saturating early while semantic properties develop more slowly. The authors release a new suite of models, NovoMolGen, that achieves state-of-the-art results in drug discovery tasks, highlighting a disconnect between representation learning and downstream design and calling for new pretraining paradigms that target chemical semantics.
By Roshan Balaji, Kamran Chitsaz, Quentin Fournier, Nirav Pravinbhai Bhatt, Sarath Chandar
arXiv:2607. 03007v1 Announce Type: cross Abstract: Recent advances in molecular large language models have led to strong performance on molecular understanding and generation tasks, yet these gains often come without reliable structural grounding.
By Wenda Wang, Jinjia Feng, Zhewei Wei
arXiv:2606. 03057v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for molecular tasks, but it remains unclear which molecular representation to use.
By Arun Raja, Garrett M. Morris, Kian Ming A. Chai
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
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.
MolSC is a new dataset of 181,000 substituent-level examples that captures how attaching specific substituents to molecular scaffolds changes properties such as bioactivity and physicochemical descriptors. The authors also provide MolSC-Bench, a held‑out benchmark of 1,541 examples that are disjoint from MolSC at scaffold, substituent, and molecule levels. Experiments show that training molecular large language models on MolSC markedly improves their ability to predict substituent contributions, outperforming existing models on a range of downstream chemistry tasks.
By Hyuntae Park, Sooyeon Kim, Jiwon Park, SangKeun Lee
arXiv:2608. 10480v1 Announce Type: new Abstract: Large language models (LLMs) are widely applied across chemical tasks, such as molecular property prediction, which underpins drug discovery.
By Junwoo Park, Minyoung Shin, Cheol Soon Lee, Sujee Lee
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
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.
By David Sulu, Lorenzo Di Fruscia, Jana M. Weber
arXiv:2606. 12113v1 Announce Type: cross Abstract: Transformer-based language models for SMILES strings suffer from a locality gap: standard character-level tokenization fragments chemically meaningful motifs, forcing models to repeatedly learn local syntax at the expense of long-range dependencies.
By Xinni Zhang, Zijing Liu, He Cao, Yu Li, Irwin King
arXiv:2606. 05693v1 Announce Type: new Abstract: Large language models (LLMs) have shown promise for molecular property prediction, but their ability to reason over chemical structures remains limited, as molecular representations such as SMILES differ substantially from the natural language on which LLMs are primarily trained.
By Joey Chan, Wonbin Kweon, Ashley Shin, Niharika Bhattacharjee, Pengcheng Jiang, Yue Guo, Jiawei Han
arXiv:2608. 06956v1 Announce Type: new Abstract: Generative models for matter are often evaluated as samplers over output representations, and their latent spaces are commonly used as proxies for navigating chemical space.
By Raul Ortega-Ochoa, Tejs Vegge, Jens S. Bakander, Luis Mantilla Calderon, Alan Aspuru-Guzik, Tonio Buonassisi