arXiv cs.AI· Zirui Song, Zheng Lu, Mingqi Gao, Qinlei Xie, Wanqi Zhong, Hanwen Cui, Lijie Wang, Chong Luo, Bei Liu, Yiming Li, Xiuying Chen·· 9 小时前AI 评分29
Causal-Plan-Bench:用因果规划评测物理具身推理,GPT-6-astra 仅得 43.04
Planning Takes More Than Token Prediction: Causal Plan for Benchmarking and Building Physically Grounded Embodied Reasoners
AI 导读
研究团队提出 Causal-Plan-Bench,一个覆盖四个因果维度的高保真诊断基准,并构建百万级因果推理语料 Causal-Plan-1M。
来源:arXiv cs.AI · arxiv.org