arXiv subjects

Hyunsoo Park

Publications and source records attributed to Hyunsoo Park.

7 recordsLinked to original sources

arXiv2026

Chemical filters for ultra-high-throughput materials screening and generation

Generative artificial intelligence is rapidly transforming materials design by enabling de novo exploration of immense chemical spaces. Yet a large proportion of AI-generated compositions remain implausible, violating established chemical principles, which limits the reliability and interpretability of generative materials design. Here, we introduce a chemical validity operator that recasts heuristic chemical rules as a configurable algorithmic prior for evaluating and guiding generative materials discovery. Built on the open-source SMACT package, a data-informed oxidation-state model exposes tunable thresholds, allowing users to interpolate continuously between permissive and conservative chemical constraints, while supporting both exploratory and conservative materials-design workflows. Benchmarking six state-of-the-art generative models for inorganic crystals shows that most reproduce stoichiometry but under-represent realistic oxidation-state combinations, and that filtering removes compositions reliant on rarely observed oxidation states while preserving low-energy compounds near the convex hull. Beyond screening, the same operator can also serve as a reinforcement-learning reward, steering a latent diffusion model towards chemically grounded compositions. By encoding chemical heuristics and observations, this work establishes a foundation for oxidation-state-aware generative models.

cond-mat.mtrl-sci↗

arXiv2025

Crystalyse: a multi-tool agent for materials design

We present Crystalyse, an open, provenance-enforced scientific agent for computational materials design of inorganic crystals that orchestrates tools for compositional screening, crystal structure generation, and machine-learning force-field evaluation. Crystalyse offers three operating modes to trade exploration speed against validation depth: creative (rapid query), adaptive (context-aware routing) and rigorous (comprehensive checks). We release the underlying source code and evaluation scripts to enable plug-and-play use and development. In demonstrations on quaternary oxide exploration, sodium-ion cathode design, and lead-free indoor photovoltaic candidate generation, the agent integrates chemical compound generation with fast stability and property filters. Under adversarial testing, provenance enforcement eliminated material-property hallucinations (a broad adversarial suite pass rate reached 86% from a 57% baseline). Crystalyse provides an agentic artificial intelligence system that can complement existing materials design pipelines, assisting in hypothesis generation while preserving transparency and reproducibility.

cond-mat.mtrl-sci↗

arXiv2025

Guiding Generative Models to Uncover Diverse and Novel Crystals via Reinforcement Learning

Discovering functional crystalline materials entails navigating an immense combinatorial design space. While recent advances in generative artificial intelligence have enabled the sampling of chemically plausible compositions and structures, a fundamental challenge remains: the objective misalignment between likelihood-based sampling in generative modelling and targeted focus on underexplored regions where novel compounds reside. Here, we introduce a reinforcement learning framework that guides latent denoising diffusion models toward diverse and novel, yet thermodynamically viable crystalline compounds. Our approach integrates group relative policy optimisation with verifiable, multi-objective rewards that jointly balance creativity, stability, and diversity. Beyond de novo generation, we demonstrate enhanced property-guided design that preserves chemical validity, while targeting desired functional properties. This approach establishes a modular foundation for controllable AI-driven inverse design that addresses the novelty-validity trade-off across scientific discovery applications of generative models.

cs.LG↗

arXiv2025

Continuous SUN (Stable, Unique, and Novel) Metric for Generative Modeling of Inorganic Crystals

To address pressing scientific challenges such as climate change, increasingly sophisticated generative models are being developed to efficiently sample the large chemical space of potential functional materials. The proliferation of these models has necessitated the establishment of rigorous evaluation metrics. While uniqueness (U), novelty (N), and stability (S) of samples serve as standard metrics, their current formulations show several limitations. U and N rely on binary comparisons of crystals, rendering them dependent on heuristic thresholds, incapable of quantifying the degree of similarity, sensitive to atomic coordinate perturbations, and not invariant to sample permutation. Similarly, the binary assessment of S risks a premature exclusion of marginally unstable yet potentially novel candidates. These limitations are addressed by making the aforementioned metrics continuous. Furthermore, we integrate them into a unified metric ``continuous SUN" (cSUN), which offers a smoother score distribution and greater tunability than the conventional binary SUN metric. Experimental results demonstrate that our continuous metrics provide granular insights into sample distributions and facilitate the identification of the most promising candidates. Finally, the use of cSUN as a reward signal in reinforcement learning is explored, showing that its adjustable weighting scheme effectively mitigates reward hacking and avoids local minima.

cs.LG↗

arXiv2025

MLIP Arena: Advancing Fairness and Transparency in Machine Learning Interatomic Potentials via an Open, Accessible Benchmark Platform

Machine learning interatomic potentials (MLIPs) have revolutionized molecular and materials modeling, but existing benchmarks suffer from data leakage, limited transferability, and an over-reliance on error-based metrics tied to specific density functional theory (DFT) references. We introduce MLIP Arena, a benchmark platform that evaluates force field performance based on physics awareness, chemical reactivity, stability under extreme conditions, and predictive capabilities for thermodynamic properties and physical phenomena. By moving beyond static DFT references and revealing the important failure modes of current foundation MLIPs in real-world settings, MLIP Arena provides a reproducible framework to guide the next-generation MLIP development toward improved predictive accuracy and runtime efficiency while maintaining physical consistency. The Python package and online leaderboard are available at https://github.com/atomind-ai/mlip-arena.

physics.chem-ph↗

arXiv2024

HandNeRF: Learning to Reconstruct Hand-Object Interaction Scene from a Single RGB Image

This paper presents a method to learn hand-object interaction prior for reconstructing a 3D hand-object scene from a single RGB image. The inference as well as training-data generation for 3D hand-object scene reconstruction is challenging due to the depth ambiguity of a single image and occlusions by the hand and object. We turn this challenge into an opportunity by utilizing the hand shape to constrain the possible relative configuration of the hand and object geometry. We design a generalizable implicit function, HandNeRF, that explicitly encodes the correlation of the 3D hand shape features and 2D object features to predict the hand and object scene geometry. With experiments on real-world datasets, we show that HandNeRF is able to reconstruct hand-object scenes of novel grasp configurations more accurately than comparable methods. Moreover, we demonstrate that object reconstruction from HandNeRF ensures more accurate execution of downstream tasks, such as grasping and motion planning for robotic hand-over and manipulation. Homepage: https://samsunglabs.github.io/HandNeRF-project-page/

cs.CV↗

arXiv2021

Mining Insights on Metal-Organic Framework Synthesis from Scientific Literature Texts

Identifying optimal synthesis conditions for metal-organic frameworks (MOFs) is a major challenge that can serve as a bottleneck for new materials discovery and development. Trial-and-error approach that relies on a chemist's intuition and knowledge has limitations in efficiency due to the large MOF synthesis space. To this end, 47,187 number of MOF were data mined using our in-house developed code to extract their synthesis information in 28,565 MOF papers. The joint machine learning/rule-based algorithm yields an average F1 score of 90.3 % across different synthesis parameters (i.e. metal precursors, organic precursors, solvents, temperature, time, composition). From this data set, a PU learning algorithm was developed to predict synthesis of a given MOF material using synthesis conditions as inputs, and this algorithm successfully predicted successful synthesis in 83.1 % of the synthesized data in the test set. Finally, our model correctly predicted three amorphous MOFs (with their representative experimental synthesis condition) as having low synthesizability scores while the counterpart crystalline MOFs showed high synthesizability scores. Our results show that big data extracted from the texts of MOF papers can be used to rationally predict synthesis conditions for these materials, which can accelerate the speed in which new MOFs are synthesized.

cond-mat.mtrl-sci↗