Scientific large language models: A survey on biological & chemical domains

Q Zhang, K Ding, T Lv, X Wang, Q Yin, Y Zhang… - ACM Computing …, 2024 - dl.acm.org
Large Language Models (LLMs) have emerged as a transformative power in enhancing
natural language comprehension, representing a significant stride toward artificial general …

A Hitchhiker's Guide to Geometric GNNs for 3D Atomic Systems

A Duval, SV Mathis, CK Joshi, V Schmidt… - arxiv preprint arxiv …, 2023 - arxiv.org
Recent advances in computational modelling of atomic systems, spanning molecules,
proteins, and materials, represent them as geometric graphs with atoms embedded as …

DPA-2: a large atomic model as a multi-task learner

D Zhang, X Liu, X Zhang, C Zhang, C Cai… - npj Computational …, 2024 - nature.com
The rapid advancements in artificial intelligence (AI) are catalyzing transformative changes
in atomic modeling, simulation, and design. AI-driven potential energy models have …

Where did the gap go? reassessing the long-range graph benchmark

J Tönshoff, M Ritzert, E Rosenbluth… - arxiv preprint arxiv …, 2023 - arxiv.org
The recent Long-Range Graph Benchmark (LRGB, Dwivedi et al. 2022) introduced a set of
graph learning tasks strongly dependent on long-range interaction between vertices …

MolE: a foundation model for molecular graphs using disentangled attention

O Méndez-Lucio, CA Nicolaou, B Earnshaw - Nature Communications, 2024 - nature.com
Abstract Models that accurately predict properties based on chemical structure are valuable
tools in the chemical sciences. However, for many properties, public and private training sets …

DPA-2: Towards a universal large atomic model for molecular and material simulation

D Zhang, X Liu, X Zhang, C Zhang, C Cai, H Bi… - arxiv preprint arxiv …, 2023 - arxiv.org
The rapid development of artificial intelligence (AI) is driving significant changes in the field
of atomic modeling, simulation, and design. AI-based potential energy models have been …

DIMAT: Decentralized Iterative Merging-And-Training for Deep Learning Models

N Saadati, M Pham, N Saleem… - Proceedings of the …, 2024 - openaccess.thecvf.com
Recent advances in decentralized deep learning algorithms have demonstrated cutting-
edge performance on various tasks with large pre-trained models. However a pivotal …

Generative ai in medicine

D Shanmugam, M Agrawal, R Movva, IY Chen… - arxiv preprint arxiv …, 2024 - arxiv.org
The increased capabilities of generative AI have dramatically expanded its possible use
cases in medicine. We provide a comprehensive overview of generative AI use cases for …

Graphfm: A scalable framework for multi-graph pretraining

D Lachi, M Azabou, V Arora, E Dyer - arxiv preprint arxiv:2407.11907, 2024 - arxiv.org
Graph neural networks are typically trained on individual datasets, often requiring highly
specialized models and extensive hyperparameter tuning. This dataset-specific approach …

Reducing the cost of quantum chemical data by backpropagating through density functional theory

A Mathiasen, H Helal, P Balanca, A Krzywaniak… - arxiv preprint arxiv …, 2024 - arxiv.org
Density Functional Theory (DFT) accurately predicts the quantum chemical properties of
molecules, but scales as $ O (N_ {\text {electrons}}^ 3) $. Sch\" utt et al.(2019) successfully …