通过跨模态知识转移增强基于组成的材料性能预测
Enhancing composition-based materials property prediction by cross-modal knowledge transfer
AI总结:
提出跨模态知识转移的通用方法,通过隐式和显式两种方案增强基于组成的材料性能预测,在多项基准任务中达到SOTA,并利用博弈论方法提升模型可解释性。
摘要:
晶体图神经网络广泛应用于建模实验合成的化合物及合成性未知的假设材料。相比之下,与结构无关的预测算法允许探索化学空间中以前无法进入的领域。在此,我们提出了一种通过跨模态知识转移来增强基于组成的材料性能预测的通用方法。提出了两种方案:隐式转移涉及在多模态嵌入上预训练化学语言模型,而显式转移则建议生成晶体结构并实现结构感知预测器。所提出的方法在 LLM4Mat-Bench 和 MatBench 任务上进行了基准测试,在32个案例中的25个取得了最先进的性能。此外,我们展示了化学语言模型的另一建模方面——可解释性——如何受益于博弈论方法的应用,该方法能够纳入高阶特征交互。
英文摘要:
Crystal graph neural networks are widely applicable in modeling experimentally synthesized compounds and hypothetical materials with unknown synthesizability. In contrast, structure-agnostic predictive algorithms allow exploring previously inaccessible domains of chemical space. Here we present a universal approach for enhancing composition-based materials property prediction by means of cross-modal knowledge transfer. Two formulations are proposed: implicit transfer involves pretraining chemical language models on multimodal embeddings, whereas explicit transfer suggests generating crystal structures and implementing structure-aware predictors. The proposed approaches were benchmarked on LLM4Mat-Bench and MatBench tasks, achieving state-of-the-art performance in 25 out of 32 cases. In addition, we demonstrated how another modeling aspect of chemical language models - interpretability - benefits from applying a game-theoretic approach, which is able to incorporate high-order feature interactions.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。