TransPolymer：一种用于聚合物性质预测的基于Transformer的语言模型
TransPolymer: a Transformer-based language model for polymer property predictions
AI总结：
该研究提出基于Transformer的聚合物性质预测语言模型TransPolymer，通过化学感知分词器学习序列表征，结合掩码语言建模预训练，在10个基准上取得更优性能，为理性聚合物设计提供新工具。
摘要：
准确高效地预测聚合物性质在聚合物设计中具有重要意义。传统方法需要开展成本高昂且耗时的实验或模拟来评估聚合物的功能。近年来，配备自注意力机制的Transformer模型在自然语言处理领域展现出优异性能，但这类方法尚未在聚合物科学中得到研究。本文提出TransPolymer，一种用于聚合物性质预测的基于Transformer的语言模型。我们设计的具备化学感知能力的聚合物分词器（polymer tokenizer）可实现从聚合物序列中学习表征。在10个聚合物性质预测基准上开展的严格实验证明，TransPolymer具有更优性能。此外，我们发现TransPolymer可通过掩码语言建模（Masked Language Modeling）在大规模无标签数据集上进行预训练，从而提升性能。实验结果进一步表明，自注意力机制在聚合物序列建模中发挥着重要作用。我们认为该模型是一种极具潜力的计算工具，可推动理性聚合物设计，并从数据科学视角理解结构-性质关系。
英文摘要：
Accurate and efficient prediction of polymer properties is of great significance in polymer design. Conventionally, expensive and time-consuming experiments or simulations are required to evaluate polymer functions. Recently, Transformer models, equipped with self-attention mechanisms, have exhibited superior performance in natural language processing. However, such methods have not been investigated in polymer sciences. Herein, we report TransPolymer, a Transformer-based language model for polymer property prediction. Our proposed polymer tokenizer with chemical awareness enables learning representations from polymer sequences. Rigorous experiments on ten polymer property prediction benchmarks demonstrate the superior performance of TransPolymer. Moreover, we show that TransPolymer benefits from pretraining on large unlabeled dataset via Masked Language Modeling. Experimental results further manifest the important role of self-attention in modeling polymer sequences. We highlight this model as a promising computational tool for promoting rational polymer design and understanding structure-property relationships from a data science view.
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。
