ibm-research/acp_bench
收藏Hugging Face2025-04-16 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/ibm-research/acp_bench
下载链接
链接失效反馈官方服务:
资源简介:
ACP Bench是一个用于评估大型语言模型在动作、变化和规划方面的推理能力的基准数据集。该数据集包含7个推理任务,分布在13个领域,包括11个经典规划领域,ALFWorld和一个创新的Swap领域。
ACP Bench is a benchmark dataset for evaluating the reasoning ability of large language models about Action, Change, and Planning. The dataset consists of 7 reasoning tasks across 13 domains, including 11 classical planning domains, ALFWorld, and a novel Swap domain.
提供机构:
ibm-research



