arXiv subjects

Zhilong Song

Publications and source records attributed to Zhilong Song.

10 recordsLinked to original sources

arXiv2026

Autonomous discovery of new structure-plausibility laws for explainable and rapid crystal diagnosis and screening

Crystal generators and tool-using agents propose structures faster than density functional theory (DFT) energy and phonon calculations or experiments can assess them. Deciding which candidates merit expensive assessment is therefore the bottleneck, yet most screens test little beyond atomic overlap and give no chemical reason for failure. Here, our agents generate, test and actively refute two million candidate laws, leaving eight Plausibility Rules for Inorganic Structures (PRIS). These laws encode five mechanisms: short-range repulsion, ionic contact and packing, electrostatic balance, bond-valence conservation and crystallographic site complexity. Experimental structures satisfy our law sets at 82--99%, but satisfy Pauling's rules 2--5 together at only 6.5%. The strictest set detects 87.9% of damaged crystal structures, whereas distance cutoffs detect only 1.6--3.2%. PRIS plausibility is linearly correlated with synthesizability, so the PRIS-derived synthesis score (PSS) explainably screens 83.7% of hard-to-synthesize structures while retaining 80.7% of experimental structures. In a property-conditioned inverse-design run, PRIS and PSS can reduce the DFT validation queue by up to 67.3% and keep 99.2% of the candidates whose DFT-validated bulk moduli reach the design target. Beyond screening, PRIS explains why GNoME remains enriched in rare low-symmetry structures and reveals how wrong-element assignments in falsified crystal reports hide behind plausible coordinates. PRIS moves screening from a pass-or-fail verdict to a chemical reason for failure, showing that autonomous agents can discover, by active refutation, physicochemical laws that guide calculations and experiments.

cond-mat.mtrl-sci↗

arXiv2026

Physics-grounded generative design of inherently stable, novel and controllable crystal structures

Generative inverse design is reshaping the discovery of functional crystalline materials. Yet current generative models face challenges in simultaneously achieving stability, novelty, and precise controllability in a single trained model. We address these challenges with a key physical insight: the diversity of crystals is governed by their crystallographic information (CI), namely composition, space group and lattice, whereas only a few stable atomic configurations remain once the CI is fixed. Built on this insight, we introduce SCGEN (stable and controllable crystal structure generation), a physics-grounded generative model with two components. A variational autoencoder samples diverse, physically plausible CI, and a symmetry- and Wyckoff-position-constrained optimizer locates stable atomic positions via universal machine-learning potentials. Benchmarked on roughly two million structures, SCGEN reaches state-of-the-art stability while preserving comparable novelty, and it satisfies any specified composition, space group, lattice or joint constraint with 100% success and no task-specific retraining. Applied to photocatalytic water splitting, property-guided optimization with SCGEN generates 200,000 candidate structures and identify top 22 stable, active and synthesizable photocatalysts. By decoupling CI generation from coordinate optimization, SCGEN establishes a physics-grounded inverse-design paradigm that yields synthesis-ready crystals on demand, rather than structures requiring post hoc repair, relaxation, or retraining.

cond-mat.mtrl-sci↗

arXiv2026

Autonomous heterogeneous catalyst discovery with a self-evolving multi-agent digital twin

Theoretical heterogeneous catalysis promises rapid catalyst discovery, yet computational and machine-learning predictions often deviate from experiment and stay confined to narrow material families, for want of a faithful, condition-aware catalytic simulator. We present CatDT (Catalysis Digital Twin), a self-evolving multi-agent system that builds an autonomous digital twin of a working catalyst, unifying gas-solid and liquid-solid modeling. From only a bulk crystal and a natural-language reaction description, eight specialized agents and 27 scientific tools predict stable facets, reconstruct working surfaces, enumerate and rank reaction pathways, locate transition states, and compute kinetics in 5-30 min on a single GPU. Two innovations address the hardest steps: UniMech finds dominant pathways for novel materials at over $10^3\times$ lower cost than exhaustive enumeration by fusing agent-guided proposals with energy-cached graph search, and a memory-augmented reinforcement loop raises barrier-calculation success from 41% to 84% across 600 catalytic surfaces. Across seven gas-solid benchmarks -- stepped metals, single-atom catalysts, ordered intermetallics, vacancy-rich 2D sulfides and carbides, and a strong-metal--support-interaction (SMSI) interface -- every CatDT prediction lies within 0.5-2 times experiment over four orders of magnitude. For propane dehydrogenation, CatDT independently discovers non-precious candidates rivaling the Pt-based industrial benchmark, with a proposed Ni@ZrO$_2$ SMSI overlayer reaching a simulated TOF of $1.63~\text{s}^{-1}$ at $\sim$100% selectivity. More broadly, the decisive factor for a faithful catalyst digital twin -- or any multi-stage scientific simulator -- is not raw LLM capability but the engineered harness around it: deterministic tools, persistent memory, and verified self-improvement that compound across models, tools, and runs.

cond-mat.mtrl-sci↗

arXiv2026

LoViF 2026 The First Challenge on Holistic Quality Assessment for 4D World Model (PhyScore)

This paper reports on the LoViF 2026 PhyScore challenge, a competition on holistic quality assessment of world-model-generated videos across both 2D and 4D generation settings. The challenge is motivated by a central gap in current evaluation practice: perceptual quality alone is insufficient to judge whether generated dynamics are physically plausible, temporally coherent, and consistent with input conditions. Participants are required to build a metric that jointly predicts four dimensions, i.e., Video Quality, Physical Realism, Condition-Video Alignment, and Temporal Consistency. Depart from that, participants also need to localize physical anomaly timestamps for fine-grained diagnosis. The benchmark dataset contains 1,554 videos generated by seven representative world generative models, organized into three tracks (text-2D, image-to-4D, and video-to-4D) and spanning 26 categories. These categories explicitly cover physics-relevant scenarios, including dynamics, optics, and thermodynamics, together with diverse real-world and creative content. To ensure label reliability, scores and anomaly timestamps are produced through trained human annotation with an additional automated quality-control pass. Evaluation is based on both score prediction and anomaly localization, with a composite protocol that combines TimeStamp_IOU and SRCC/PLCC. This report summarizes the challenge design and provides method-level insights from submitted solutions.

cs.CV↗

arXiv2025

T2MAT (text-to-materials): A universal agent for generating material structures with goal properties from a single sentence

Artificial Intelligence-Generated Content (AIGC)-content autonomously produced by AI systems without human intervention-has significantly boosted efficiency across various fields. However, AIGC in material science faces challenges in efficiently discovering novel materials that surpass existing databases, while simultaneously addressing the invariance and stability of crystal structures. To address these challenges, we develop T2MAT (text-to-material), a comprehensive agent processing from a user-input sentence to inverse design material structures with goal properties beyond the existing database via globally exploring chemical space, followed by an entirely automated workflow of first-principles validation. Furthermore, we propose CGTNet (Crystal Graph Transformer NETwork), a graph neural network model that captures long-range interactions, to enhance the accuracy and data utilization efficiency of property prediction and thereby strengthen the reliability of inverse design. Through these contributions, T2MAT minimizes the dependency on human expertise and significantly improves the efficiency of discovering novel, high-performance functional materials, offering a robust way toward more autonomous materials design.

cond-mat.mtrl-sci↗

arXiv2025

LLM-Feynman: Leveraging Large Language Models for Universal Scientific Formula and Theory Discovery

Distilling underlying principles from data has historically driven scientific breakthroughs. However, conventional data-driven machine learning often produces complex models that lack interpretability and generalization due to insufficient domain expertise. Here, we present LLM-Feynman, a novel framework that leverages large language models (LLMs) alongside systematic optimization to derive concise, interpretable formulas from data and domain knowledge. Our method integrates automated feature engineering, LLM-guided symbolic regression with self-evaluation, and Monte Carlo tree search to enhance formula discovery and clarity. The embedding of domain knowledge simplifies the formula, while self-evaluation based on this knowledge further minimizes prediction errors, surpassing conventional symbolic regression in accuracy and interpretability. Our LLM-Feynman successfully rediscovered over 90% of fundamental physical formulas and demonstrated its efficacy in key materials science applications, including classification of two-dimensional material and perovskite synthesizability and determination of the Green's function and screened Coulomb interaction bandgaps, and prediction of ionic conductivity in lithium solid-state electrolytes. By transcending mere data fitting through the integration of deep domain knowledge, this LLM-Feynman offers a transformative paradigm for the automated discovery of generalizable scientific formulas and theories across disciplines.

cond-mat.mtrl-sci↗

arXiv2025

GUI-Robust: A Comprehensive Dataset for Testing GUI Agent Robustness in Real-World Anomalies

The development of high-quality datasets is crucial for benchmarking and advancing research in Graphical User Interface (GUI) agents. Despite their importance, existing datasets are often constructed under idealized conditions, overlooking the diverse anomalies frequently encountered in real-world deployments. To address this limitation, we introduce GUI-Robust, a novel dataset designed for comprehensive GUI agent evaluation, explicitly incorporating seven common types of anomalies observed in everyday GUI interactions. Furthermore, we propose a semi-automated dataset construction paradigm that collects user action sequences from natural interactions via RPA tools and then generate corresponding step and task descriptions for these actions with the assistance of MLLMs. This paradigm significantly reduces annotation time cost by a factor of over 19 times. Finally, we assess state-of-the-art GUI agents using the GUI-Robust dataset, revealing their substantial performance degradation in abnormal scenarios. We anticipate that our work will highlight the importance of robustness in GUI agents and inspires more future research in this direction. The dataset and code are available at https://github.com/chessbean1/GUI-Robust..

cs.AI↗

arXiv2024

Is Large Language Model All You Need to Predict the Synthesizability and Precursors of Crystal Structures?

Accessing the synthesizability of crystal structures is pivotal for advancing the practical application of theoretical material structures designed by machine learning or high-throughput screening. However, a significant gap exists between the actual synthesizability and thermodynamic or kinetic stability, which is commonly used for screening theoretical structures for experiments. To address this, we develop the Crystal Synthesis Large Language Models (CSLLM) framework, which includes three LLMs for predicting the synthesizability, synthesis methods, and precursors. We create a comprehensive synthesizability dataset including 140,120 crystal structures and develop an efficient text representation method for crystal structures to fine-tune the LLMs. The Synthesizability LLM achieves a remarkable 98.6% accuracy, significantly outperforming traditional synthesizability screening based on thermodynamic and kinetic stability by 106.1% and 44.5%, respectively. The Methods LLM achieves a classification accuracy of 91.02%, and the Precursors LLM has an 80.2% success rate in predicting synthesis precursors. Furthermore, we develop a user-friendly graphical interface that enables automatic predictions of synthesizability and precursors from uploaded crystal structure files. Through these contributions, CSLLM bridges the gap between theoretical material design and experimental synthesis, paving the way for the rapid discovery of novel and synthesizable functional materials.

cond-mat.mtrl-sci↗

arXiv2024

Inverse Design of Promising Alloys for Electrocatalytic CO$_2$ Reduction via Generative Graph Neural Networks Combined with Bird Swarm Algorithm

Directly generating material structures with optimal properties is a long-standing goal in material design. One of the fundamental challenges lies in how to overcome the limitation of traditional generative models to efficiently explore the global chemical space rather than a small localized space. Herein, we develop a framework named MAGECS to address this dilemma, by integrating the bird swarm algorithm and supervised graph neural network to effectively navigate the generative model in the immense chemical space towards materials with target properties. As a demonstration, MAGECS is applied to design compelling alloy electrocatalysts for CO$_2$ reduction reaction (CO$_2$RR) and works extremely well. Specifically, the chemical space of CO$_2$RR is effectively explored, where over 250,000 promising structures with high activity have been generated and notably, the proportion of desired structures is 2.5-fold increased. Moreover, five predicted alloys, i.e., CuAl, AlPd, Sn$_2$Pd$_5$, Sn$_9$Pd$_7$, and CuAlSe$_2$ are successfully synthesized and characterized experimentally, two of which exhibit about 90% Faraday efficiency of CO$_2$RR, and CuAl achieved 76% efficiency for C$_2$ products. This pioneering application of inverse design in CO$_2$RR catalysis showcases the potential of MAGECS to dramatically accelerate the development of functional materials, paving the way for fully automated, artificial intelligence-driven material design.

cond-mat.mtrl-sci↗

arXiv2019

Symbolic Regression Discovery of New Perovskite Catalysts with High Oxygen Evolution Reaction Activity

Symbolic regression (SR) is an emerging method for building analytical formulas to find models that best fit data sets. Here, SR was used to guide the design of new oxide perovskite catalysts with improved oxygen evolution reaction (OER) activities. An unprecedentedly simple descriptor, μ/t, where μ and t are the octahedral and tolerance factors, respectively, was identified, which accelerated the discovery of a series of new oxide perovskite catalysts with improved OER activity. We successfully synthesized five new oxide perovskites and characterized their OER activities. Remarkably, four of them, Cs0.4La0.6Mn0.25Co0.75O3, Cs0.3La0.7NiO3, SrNi0.75Co0.25O3, and Sr0.25Ba0.75NiO3, outperform the current state-of-the-art oxide perovskite catalyst, Ba0.5Sr0.5Co0.8Fe0.2O3 (BSCF). Our results demonstrate the potential of SR for accelerating data-driven design and discovery of new materials with improved properties.

cond-mat.mtrl-sci↗