LAMDA-NeSy/chinatravel_neurips25submission

ChinaTravel是一个精心设计的基准数据集，旨在为多日多景点旅行规划中的语言代理提供一个全面且可扩展的评估框架。数据集包含合成数据、验证数据和测试数据，以及针对不同偏好的配置。每个数据分割包含不同数量的查询，难度各不相同，支持中文和英文两种语言。ChinaTravel is a meticulously designed benchmark that provides a comprehensive and scalable evaluation framework for language agents in multi-day multi-POI travel planning. The dataset includes synthetic, validation, and test data, as well as configurations tailored to different preferences. Each data split contains a different number of queries with varying difficulty levels, and supports both Chinese and English languages.