QE-Catalytic: A Graph-Language Multimodal Base Model for Relaxed-Energy Prediction in Catalytic Adsorption

• URL: http://arxiv.org/abs/2512.20084v1
• Date: Tue, 23 Dec 2025 06:27:30 GMT
• Title: QE-Catalytic: A Graph-Language Multimodal Base Model for Relaxed-Energy Prediction in Catalytic Adsorption
• Authors: Yanjie Li, Jian Xu, Xueqing Chen, Lina Yu, Shiming Xiang, Weijun Li, Cheng-lin Liu,
• Abstract summary: We propose QE-Catalytic, a multimodal framework that couples a large language model with an E(3)-equivariant graph Transformer.<n>During prediction, QE-Catalytic jointly leverages three-dimensional structures and structured configuration text, and injects 3D geometric information'' into the language channel.<n>On OC20, QE-Catalytic reduces the MAE of relaxed adsorb energy from 0.713eV to 0.486eV, and consistently outperforms baseline models.
• Score: 44.77883047868218
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Adsorption energy is a key descriptor of catalytic reactivity. It is fundamentally defined as the difference between the relaxed total energy of the adsorbate-surface system and that of an appropriate reference state; therefore, the accuracy of relaxed-energy prediction directly determines the reliability of machine-learning-driven catalyst screening. E(3)-equivariant graph neural networks (GNNs) can natively operate on three-dimensional atomic coordinates under periodic boundary conditions and have demonstrated strong performance on such tasks. In contrast, language-model-based approaches, while enabling human-readable textual descriptions and reducing reliance on explicit graph -- thereby broadening applicability -- remain insufficient in both adsorption-configuration energy prediction accuracy and in distinguishing ``the same system with different configurations,'' even with graph-assisted pretraining in the style of GAP-CATBERTa. To this end, we propose QE-Catalytic, a multimodal framework that deeply couples a large language model (\textbf{Q}wen) with an E(3)-equivariant graph Transformer (\textbf{E}quiformer-V2), enabling unified support for adsorption-configuration property prediction and inverse design on complex catalytic surfaces. During prediction, QE-Catalytic jointly leverages three-dimensional structures and structured configuration text, and injects ``3D geometric information'' into the language channel via graph-text alignment, allowing it to function as a high-performance text-based predictor when precise coordinates are unavailable, while also autoregressively generating CIF files for target-energy-driven structure design and information completion. On OC20, QE-Catalytic reduces the MAE of relaxed adsorption energy from 0.713~eV to 0.486~eV, and consistently outperforms baseline models such as CatBERTa and GAP-CATBERTa across multiple evaluation protocols.

Related papers

• A Cross-Domain Graph Learning Protocol for Single-Step Molecular Geometry Refinement [0.47664901548798794]
  GeoOpt-Net is a multi-branch SE(3)-equivariant geometry refinement network that predicts DFT-quality structures at the B3LYP/TZVP level of theory in a single forward pass.<n>GeoOpt-Net generates initial guesses intrinsically compatible with DFT convergence criteria, yielding nonzero All-YES'' convergence rates.<n>These results establish GeoOpt-Net as a scalable, physically consistent geometry refinement framework that enables efficient acceleration of DFT-based quantum-chemical predictions.
  arXiv  Detail & Related papers  (2026-01-30T08:55:46Z)
• Accelerating High-Throughput Catalyst Screening by Direct Generation of Equilibrium Adsorption Structures [0.0]
  We present DBCata, a deep generative model that integrates a periodic Brownian-bridge framework with an equivariant graph neural network to establish a low-dimensional transition manifold between unrelaxed and DFT-relaxed structures.<n>Upon training, DBCata effectively generates high-fidelity, interatomic distance mean absolute error (DMAE) of 0.035 text on the Catalysis-Hub dataset.<n>The corresponding DFT accuracy can be improved within 0.1 eV in 94% of instances by identifying and refining anomalous predictions.
  arXiv  Detail & Related papers  (2025-12-17T09:26:58Z)
• ReDiSC: A Reparameterized Masked Diffusion Model for Scalable Node Classification with Structured Predictions [64.17845687013434]
  We propose ReDiSC, a structured diffusion model for structured node classification.<n>We show that ReDiSC achieves superior or highly competitive performance compared to state-of-the-art GNN, label propagation, and diffusion-based baselines.<n> Notably, ReDiSC scales effectively to large-scale datasets on which previous structured diffusion methods fail due to computational constraints.
  arXiv  Detail & Related papers  (2025-07-19T04:46:53Z)
• Generating Generalised Ground-State Ansatzes from Few-Body Examples [0.0]
  We introduce a method that generates ground-state ansatzes for quantum many-body systems.<n>We find analytically tractable ansatzes with a degree of universality, which encode correlations, capture finite-size effects, accurately predict ground-state energies, and offer a good description of critical phenomena.
  arXiv  Detail & Related papers  (2025-03-01T13:52:57Z)
• On the importance of catalyst-adsorbate 3D interactions for relaxed energy predictions [98.70797778496366]
  We investigate whether it is possible to predict a system's relaxed energy in the OC20 dataset while ignoring the relative position of the adsorbate. We find that while removing binding site information impairs accuracy as expected, modified models are able to predict relaxed energies with remarkably decent MAE.
  arXiv  Detail & Related papers  (2023-10-10T14:57:04Z)
• EasyDGL: Encode, Train and Interpret for Continuous-time Dynamic Graph Learning [92.71579608528907]
  This paper aims to design an easy-to-use pipeline (termed as EasyDGL) composed of three key modules with both strong ability fitting and interpretability. EasyDGL can effectively quantify the predictive power of frequency content that a model learn from the evolving graph data.
  arXiv  Detail & Related papers  (2023-03-22T06:35:08Z)
• DR-Label: Improving GNN Models for Catalysis Systems by Label Deconstruction and Reconstruction [72.20024514713633]
  We present a novel graph neural network (GNN) supervision and prediction strategy DR-Label. The strategy enhances the supervision signal, reduces the multiplicity of solutions in edge representation, and encourages the model to provide node predictions robust. DR-Label was applied to three radically distinct models, each of which displayed consistent performance enhancements.
  arXiv  Detail & Related papers  (2023-03-06T04:01:28Z)
• Multi-Task Mixture Density Graph Neural Networks for Predicting Cu-based Single-Atom Alloy Catalysts for CO2 Reduction Reaction [61.9212585617803]
  Graph neural networks (GNNs) have drawn more and more attention from material scientists. We develop a multi-task (MT) architecture based on DimeNet++ and mixture density networks to improve the performance of such task.
  arXiv  Detail & Related papers  (2022-09-15T13:52:15Z)
• Boosting Heterogeneous Catalyst Discovery by Structurally Constrained Deep Learning Models [0.0]
  Deep learning approaches such as graph neural networks (GNNs) open new opportunity to significantly extend scope for modelling novel high-performance catalysts. Here we present embedding improvement for GNN that has been modified by Voronoi tesselation. We show that a sensible choice of data can decrease the error to values above physically-based 20 meV per atom threshold.
  arXiv  Detail & Related papers  (2022-07-11T17:01:28Z)
• ExpressivE: A Spatio-Functional Embedding For Knowledge Graph Completion [78.8942067357231]
  ExpressivE embeds pairs of entities as points and relations as hyper-parallelograms in the virtual triple space. We show that ExpressivE is competitive with state-of-the-art KGEs and even significantly outperforms them on W18RR.
  arXiv  Detail & Related papers  (2022-06-08T23:34:39Z)
• RCC-GAN: Regularized Compound Conditional GAN for Large-Scale Tabular Data Synthesis [7.491711487306447]
  This paper introduces a novel generative adversarial network (GAN) for synthesizing large-scale databases. We propose a new formulation for deriving a vector incorporating both binary and discrete features simultaneously. We present a regularization scheme towards limiting unprecedented variations on its weight vectors during training.
  arXiv  Detail & Related papers  (2022-05-24T01:14:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.