DisastIR

该数据集名为DisastIR，是一个专为灾害管理定制的全面信息检索基准测试。它包含了9,600个多样化的用户查询以及超过130万个标记的查询-段落对，覆盖了48个不同的检索任务。这些任务源自六个搜索意图和八个一般灾害类别。此外，DisastIR还包括了对30个最先进的检索模型的性能评估，揭示了通用领域与灾害管理特定任务之间显著的性能差距。该数据集的规模涉及9,600个查询和超过130万个查询-段落对，其任务专注于灾害管理的信息检索。

The dataset, named DisastIR, is a comprehensive information retrieval benchmark tailored for disaster management. It contains 9,600 diverse user queries and over 1.3 million annotated query-passage pairs, covering 48 distinct retrieval tasks derived from six search intents and eight general disaster categories. Additionally, DisastIR includes performance evaluations of 30 state-of-the-art retrieval models, revealing a significant performance gap between general-domain and disaster management-specific tasks. With a scale involving 9,600 queries and over 1.3 million query-passage pairs, its tasks focus on information retrieval for disaster management.