Hyejin Jeon

Data from: Agentic programs as an emerging form of scientific software in computational materials science

Yunsung Lim, Hyejin Jeon, Jaesun Kim, Jisu Kim et autres

This repository contains the data from DeMARS package that was deployed along with the perspective paper "Agentic programs: an emerging form of scientific software in computational materials science". Contents included: runs.tar.gz: data of 100 hold-out test cases from DeMARS package icsd_id-used_for_maturation.csv: List …

Data from: Agentic programs as an emerging form of scientific software in computational materials science

Yunsung Lim, Hyejin Jeon, Jaesun Kim, Jisu Kim et autres

This repository contains the data from DeMARS package that was deployed along with the perspective paper "Agentic programs: an emerging form of scientific software in computational materials science". Contents included: runs.tar.gz: data of 100 hold-out test cases from DeMARS package icsd_id-used_for_maturation.csv: List …

Agentic programs: an emerging form of scientific software in computational materials science

Yunsung Lim, Hyejin Jeon, Jaesun Kim, Jisu Kim et autres

Computational materials science has traditionally delegated algorithmic tasks to computers while leaving scientific judgments to humans. We argue that recent LLM-based agent harnesses enable an emerging form of scientific software, agentic programs, that combine deterministic algorithms with bounded LLM-based judgment, task-specific verification, …

MDIL-SNU/SevenNet: v0.13.0

Yutack Park, jinmu.you, Gijin Kim, Jaesun0912 et autres

[0.13.0] Added reEWC fine-tuning with forgetting prevention for single-modal models: optional experience replay (rehearsal, load_memory_path, mem_batch_size, mem_ratio) and an Elastic Weight Consolidation penalty from a precomputed Fisher matrix (continue.fisher_information, continue.opt_params, continue.ewc_lambda), plus a cosineannealingwarmuplr scheduler Support OpenEquivariance Per-atom stress (atomic virial) support …

kr, us, cz, th (code pays fourni par la source)

Gestodene Accelerates Cutaneous Wound Healing via PAR1-Selective Positive Allosteric Modulation

Hyejin Jeon, Yunkyung Heo, Yechan Lee, So-Hyeon Park et autres

Protease-activated receptor 1 (PAR1), a G protein-coupled receptor, plays a central role in coordinating multiple phases of cutaneous wound healing, including hemostasis, cell proliferation, migration, and extracellular matrix remodeling. Despite its therapeutic potential, PAR1-selective positive allosteric modulators (PAMs) remain limited. Here, we …

kr (code pays fourni par la source)

Precipitate phase selection and grain boundary morphology in Cu-Ni-Si-Mn alloys: A machine-learning interatomic potential study

Aadil Fayaz Wani, Il‐Seok Jeong, Hyejin Jeon, Jaesun Kim et autres

Alloys inevitably contain interphase boundaries, whose energetics govern nucleation processes and precipitate morphology. In Cu-Ni-Si alloys, Mn addition markedly changes grain boundary (GB) precipitation behavior. While GB precipitation of stable Ni$_2$Si in Mn-free alloys is associated with degraded mechanical properties, Mn addition …

Precipitate phase selection and grain boundary morphology in Cu-Ni-Si-Mn alloys: A machine-learning interatomic potential study

Aadil Fayaz Wani, Il‐Seok Jeong, Hyejin Jeon, Jaesun Kim et autres

Alloys inevitably contain interphase boundaries, whose energetics govern nucleation processes and precipitate morphology. In Cu-Ni-Si alloys, Mn addition markedly changes grain boundary (GB) precipitation behavior. While GB precipitation of stable Ni$_2$Si in Mn-free alloys is associated with degraded mechanical properties, Mn addition …

kr, Éthiopie (code pays fourni par la source)

Reactive Machine Learning Interatomic Potentials for Chemistry and Materials Science

Jisu Kim, Hyunsung Cho, Hyejin Jeon, Jisu Jung et autres

In the era of artificial intelligence (AI), machine learning interatomic potentials (MLIPs) have revolutionized materials science and engineering, enabling large-scale and accurate atomistic modeling in various chemical systems. The performance of these potentials hinges on two key components: the model architecture, which …

kr, us (code pays fourni par la source)

Optimizing cross-domain transfer for universal machine learning interatomic potentials

Jaesun Kim, Jinmu You, Yutack Park, Yunsung Lim et autres

Accurate yet transferable machine-learning interatomic potentials are essential for accelerating materials and chemical discovery. However, many existing universal models are overfitted to narrow chemical spaces or computational protocols, limiting their reliability across diverse chemical and functional domains. Here, we introduce a transferable …

kr, pr (code pays fourni par la source)

MDIL-SNU/SevenNet: v0.12.1

Yutack Park, jinmu.you, Gijin Kim, Jaesun0912 et autres

[0.12.1] Fixed FlashTP with LAMMPS parallel in torch @alphalm4 Single-atom inference failure with ASE+flashTP, LAMMPS-Torch and LAMMPS ML-IAP @alphalm4 CI/CD bugs @CompRhys , @alphalm4 inconsistent argument namings of flashTP @jiheon-dev, @YutackPark , @alphalm4 Use ase.rattle for some unit tests for non-zero force …

kr, cn, us, th (code pays fourni par la source)

Jaesun0912/SevenNet: v0.12.1

Yutack Park, jinmu.you, Gijin Kim, Jaesun0912 et autres

SevenNet - a graph neural network interatomic potential package supporting efficient multi-GPU parallel molecular dynamics simulations.

kr, us, cn, th (code pays fourni par la source)

Jaesun0912/SevenNet: v0.12.1

Yutack Park, jinmu.you, Gijin Kim, Jaesun0912 et autres

SevenNet - a graph neural network interatomic potential package supporting efficient multi-GPU parallel molecular dynamics simulations.

kr, us, cn, th (code pays fourni par la source)