ProtT3 蛋白质文本问答数据集

ProtT3 数据集由新加坡国立大学、中国科学技术大学和北海道大学的研究团队于 2024 年联合构建，相关论文成果为「ProtT3: Protein-to-Text Generation for Text-based Protein Understanding」，并已入选 ACL 2024 。该数据集为论文研究的预训练数据集。

The ProtT3 dataset was jointly constructed in 2024 by research teams from the National University of Singapore, University of Science and Technology of China, and Hokkaido University. The associated research paper is titled "ProtT3: Protein-to-Text Generation for Text-based Protein Understanding" and has been accepted by ACL 2024. This dataset serves as the pre-training dataset for the research reported in the paper.

搜集汇总

背景与挑战

背景概述

ProtT3是一个用于蛋白质文本生成和理解的预训练数据集，由Swiss-Prot、ProteinKG25和PDB-QA三个子数据集构成，适用于蛋白质功能、位置和家族等文本描述生成任务。该数据集由多所大学联合构建，相关论文入选ACL 2024，具有较高的学术价值。