InvDesMobility: a reliability-gated first-principles feedback framework for closed-loop materials discovery

2026/06/15 by Wen-Kao Li, Ze-Feng Gao, Peng-Jie Guo +2 · 1 voice
Physics and Astronomy · Computer Science · #cond-mat.mtrl-sci #cs.AI

paper · pdf

Abstract

Inverse materials design starts from target functionality and searches for structures that can realize it. Its value in closed-loop discovery depends not only on prediction performance, but also on whether expensive first-principles results are independently validated, provenance-recorded, and admitted as feedback only when evidence is sufficient. This is especially important for composite properties such as carrier mobility, where a final scalar value hides intermediate quantities, fit quality, convergence history, and workflow assumptions. Here we present InvDesMobility, a reliability-gated first-principles feedback framework that integrates multi-agent automated DFT, evidence stratification, generative structure proposal, acquisition ranking, and auditable release. Using 516 2DMatPedia-derived candidates, the workflow produced 280 QC-passed materials and 573 retained carrier-direction seed channels after channel-level reliability gating. These records were split into two feedback objects: relaxed structures updated the generative model, while retained mobility channels trained the acquisition model and set validation priority. Over multiple iterations, InvDesMobility screened 2.4 x 10⁶ structures, submitted 102 candidates for DFT validation, and retained 86 reliability-gated generated channels across 41 formulas. Overall, the main contribution is not a fixed list of high-mobility materials, but a transferable feedback contract that makes closed-loop inverse design both useful and auditable when learning from expensive calculated properties. All source data, retained feedback records, and workflows are available at https://github.com/DreamLufei/invDesMobility, with an accompanying evidence website at https://dreamlufei.github.io/invDesMobility/.

Citations

• PhononBench:A Large-Scale Phonon-Based Benchmark for Dynamical Stability in Crystal Generation
• InvDesFlow-AL: active learning-based workflow for inverse design of functional materials
• InvDesFlow: An AI-driven materials inverse design workflow to explore possible high-temperature superconductors
• Autonomous chemical research with large language models
• An autonomous laboratory for the accelerated synthesis of inorganic materials
• Scaling deep learning for materials discovery
• Crystal Structure Prediction by Joint Equivariant Diffusion
• Inverse design of two-dimensional materials with invertible neural networks
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• Self-driving laboratory for accelerated discovery of thin-film materials
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• The Computational 2D Materials Database: High-Throughput Modeling and Discovery of Atomically Thin Crystals
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• 2D materials and van der Waals heterostructures
• Electron-phonon interactions from first principles
• The FAIR Guiding Principles for scientific data management and stewardship
• First principles phonon calculations in materials science
• High-mobility transport anisotropy and linear dichroism in few-layer black phosphorus
• First-principles study of the phonon-limited mobility in n-type single-layer MoS2
• Effect of the damping function in dispersion corrected density functional theory
• A consistent and accurateab initioparametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu
• From ultrasoft pseudopotentials to the projector augmented-wave method
• Generalized Gradient Approximation Made Simple
• Efficient iterative schemes forab initiototal-energy calculations using a plane-wave basis set

Discussions

• InvDesMobility: a reliability-gated first-principles feedback framework for closed-loop materials discovery https://arxiv.org/abs/2606.16133 [bsky, 0 points, 0 comments]

Related

• AIMatDesign: Knowledge-Augmented Reinforcement Learning for Inverse Materials Design under Data Scarcity
• Towards Automated Discovery: A Review of Generative Models, Multimodal Learning and Closed-Loop Workflows in Inverse Materials Design
• Accelerating inverse materials design using generative diffusion models with reinforcement learning
• GUIDe: Generative and Uncertainty-Informed Inverse Design for On-Demand Nonlinear Functional Responses
• AI-driven materials design: a mini-review