inverse-design

This paper presents a conditional catalyst generative model based on GPT architecture, pretrained on 133 million catalyst structures, achieving 98% structural validity and enabling controllable inverse design for targeted properties such as binding energy.

This paper presents a range-aware Bayesian optimization framework that directly scores the posterior probability that a candidate satisfies a target property range, enabling discovery of diverse valid designs across multiple specifications.

Researchers from MIT present a methodology for inverse design of nuclear critical experiments using deep neural networks with a novel multigroup attention pooling architecture and gradient-based optimization to maximize neutronic similarity coefficients. The approach is applied to validate a HALEU fuel transportation cask, achieving high similarity scores for three configurations of interest.

PolyFusionAgent is a framework that combines a multimodal polymer foundation model (PolyFusion) with a tool-augmented, literature-grounded design agent (PolyAgent) for polymer property prediction and inverse design, enabling evidence-linked discovery.

Proposes CoMole, a controllable molecular generative foundation model using motif-aware graph diffusion and reinforcement learning, achieving superior controllability across materials and drug discovery benchmarks.

This paper introduces RL-Kirigami, a framework combining optimal-transport conditional flow matching and reinforcement learning to solve the inverse design problem for kirigami metamaterials, achieving high accuracy and enabling rapid laser-cut prototype fabrication.