提出文本引导扩散模型 TGDMat，通过周期性 E(3)-等变 GNN 联合生成 3D 周期性材料的原子坐标、类型和晶格结构，在结构和生成任务上超越基线并降低计算开销。
摘要：
等变扩散模型已成为生成新型晶体材料的主流方法，因为它们能够利用周期性材料结构的物理对称性。然而，当前的模型无法在一个统一的端到端扩散框架中有效学习晶体材料的原子类型、分数坐标和晶格结构的联合分布。此外，这些模型均无法在用户指定生成结构必须匹配的期望特性的现实设置下工作。在本研究中，我们提出了 TGDMat，一种专为 3D 周期性材料生成设计的新型文本引导扩散模型。该方法在每个去噪步骤中通过文本描述整合全局结构知识，同时使用周期性 E(3)-等变图神经网络（GNN）联合生成原子坐标、类型和晶格结构。在基准任务上使用流行数据集进行的大量实验表明，TGDMat 以明显优势优于现有的基线方法。值得注意的是，在结构预测任务中，仅使用一个生成样本，TGDMat 便超越了所有基线模型，凸显了文本引导扩散的重要性。此外，在生成任务中，TGDMat 超越了所有基线及其文本融合变体，展示了联合扩散范式的有效性。另外，引入文本知识在利用来自专家的真实世界文本提示时，降低了整体训练和采样的计算开销，同时提升了生成性能。
英文摘要：
Equivariant diffusion models have emerged as the prevailing approach for generating novel crystal materials due to their ability to leverage the physical symmetries of periodic material structures. However, current models do not effectively learn the joint distribution of atom types, fractional coordinates, and lattice structure of the crystal material in a cohesive end-to-end diffusion framework. Also, none of these models work under realistic setups, where users specify the desired characteristics that the generated structures must match. In this work, we introduce TGDMat, a novel text-guided diffusion model designed for 3D periodic material generation. Our approach integrates global structural knowledge through textual descriptions at each denoising step while jointly generating atom coordinates, types, and lattice structure using a periodic-E(3)-equivariant graph neural network (GNN). Extensive experiments using popular datasets on benchmark tasks reveal that TGDMat outperforms existing baseline methods by a good margin. Notably, for the structure prediction task, with just one generated sample, TGDMat outperforms all baseline models, highlighting the importance of text-guided diffusion. Further, in the generation task, TGDMat surpasses all baselines and their text-fusion variants, showcasing the effectiveness of the joint diffusion paradigm. Additionally, incorporating textual knowledge reduces overall training and sampling computational overhead while enhancing generative performance when utilizing real-world textual prompts from experts.
Diffusion-Guided Search via Exponential Tilting (DiffTilt): An Application to Falsification of Safety-Critical Systems
School of Computing and Augmented Intelligence, Arizona State University（亚利桑那州立大学计算与增强智能学院）；Toyota Motor North America, Research & Development（丰田汽车北美研发公司）
REGEN: Reference-Guided Synthetic Multivariate Time Series Generation for Forecasting
Birla AI Labs, Office of Ananya Birla（Birla AI实验室，Ananya Birla办公室）；Birla Institute of Technology and Science, Pilani（Birla理工学院与科学学院，Pilani）；Kalinga Institute of Industrial Technology, Bhubaneswar（Kalinga工业技术学院，Bhubaneswar）
JAM: Keypoint-Guided Joint Prediction after Classification-Aware Marginal Proposal for Multi-Agent Interaction
College of Computer Science and Software Engineering, Shenzhen University（深圳大学计算机科学与软件工程学院）；Carleton University（卡尔顿大学）；Shenzhen Key Laboratory of Robotics and Computer Vision, SUSTech（深圳机器人与计算机视觉重点实验室，华南理工大学）
Semantic Encoder Guided Generative Adversarial Face Ultra-Resolution Network
Lakehead University（湖首大学）；Western University（韦仕敦大学）；Okanagan College（奥肯那根学院）；Hunan Normal University（湖南师范大学）；Tianjin University（天津大学）；The University of British Columbia Okanagan（不列颠哥伦比亚大学奥肯那根分校）
