GENEVA
收藏资源简介:
GENEVA是一个用于事件参数提取(EAE)模型泛化能力评估的数据集,由加州大学洛杉矶分校计算机科学系创建。该数据集包含115种事件类型和220种参数角色,显著特点是包含大量非实体参数角色。数据集通过将FrameNet数据集转换并结合人工专家注释构建,旨在评估模型在处理有限数据和未见事件类型泛化方面的能力。GENEVA的应用领域包括知识图谱构建、问答系统等,旨在解决现有数据集在事件类型和参数角色多样性上的不足。
GENEVA is a dataset for evaluating the generalization capability of event argument extraction (EAE) models, developed by the Department of Computer Science at the University of California, Los Angeles. This dataset includes 115 event types and 220 argument roles, with a prominent characteristic being the presence of a large number of non-entity argument roles. It is constructed by converting the FrameNet dataset and combining it with manual expert annotations, aiming to assess models' capacity for generalization under limited data scenarios and to unseen event types. The application areas of GENEVA cover knowledge graph construction, question answering systems, and so on, and it is designed to address the deficiencies of existing datasets in terms of the diversity of event types and argument roles.

- 1GENEVA: Benchmarking Generalizability for Event Argument Extraction with Hundreds of Event Types and Argument Roles加州大学洛杉矶分校计算机科学系 · 2023年



