标题： MolPIF：一种用于分子生成的参数插值流模型 显示英文标题

标题： MolPIF: A Parameter Interpolation Flow Model for Molecule Generation

摘要： 深度学习在分子生成中的进展显示出加速药物发现的潜力。 贝叶斯流网络（BFNs）最近在多种化学任务中表现出色，其成功通常归因于在低方差参数空间中建模的范式。 然而，基于贝叶斯推断的策略对设计更灵活的分布变换路径施加了限制，使得适应不同的数据分布和任务需求变得具有挑战性。 此外，更简单、更高效的基于参数空间的模型的潜力尚未被探索。 为了解决这个问题，我们提出了一种名为PIF的新参数插值流模型，并提供了详细的理论基础、训练和推理过程。 然后，我们开发了MolPIF用于基于结构的药物设计，与基线方法相比，展示了其在多种指标上的优越性能。 这项工作验证了基于参数空间的生成建模范式在分子中的有效性，并为模型设计提供了新的视角。 显示英文摘要

摘要： Advances in deep learning for molecular generation show promise in accelerating drug discovery. Bayesian Flow Networks (BFNs) have recently shown impressive performance across diverse chemical tasks, with their success often ascribed to the paradigm of modeling in a low-variance parameter space. However, the Bayesian inference-based strategy imposes limitations on designing more flexible distribution transformation pathways, making it challenging to adapt to diverse data distributions and varied task requirements. Furthermore, the potential for simpler, more efficient parameter-space-based models is unexplored. To address this, we propose a novel Parameter Interpolation Flow model (named PIF) with detailed theoretical foundation, training, and inference procedures. We then develop MolPIF for structure-based drug design, demonstrating its superior performance across diverse metrics compared to baselines. This work validates the effectiveness of parameter-space-based generative modeling paradigm for molecules and offers new perspectives for model design.