Language Rearrangement
收藏arXiv2025-09-30 收录
下载链接：
https://llm-rl.github.io
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个由150,000个训练任务和1,000个测试任务组成的基准，旨在支持研究语言条件下的重新排列问题。此外，该数据集专门为研究界设计，帮助他们探索语言条件下的大规模多任务实体人工智能问题。该数据集的规模包括150,000个训练任务和1,000个测试任务，所涉及的任务类型为语言条件下的重新排列。
This dataset is a benchmark consisting of 150,000 training tasks and 1,000 test tasks, designed to support research on rearrangement problems under language conditions. Moreover, this dataset is specifically developed for the research community to help explore large-scale multi-task embodied artificial intelligence problems under language conditions. The dataset covers 150,000 training tasks and 1,000 test tasks, and all involved tasks fall into the category of rearrangement tasks under language conditions.
提供机构：
Research community
