构建安全可控的机器人任务规划评估基准，解决现实场景中的隐性约束难题。
SPOC: Safety-Aware Planning Under Partial Observability And Physical Constraints
- 设计多层级约束机制，融合部分可观测性与物理规则
- 覆盖5类家庭危险，支持状态与约束双维度在线评估
- 揭示主流大模型在隐式约束下严重缺乏安全规划能力
具身任务规划中使用大语言模型面临真实环境中的安全挑战，需同时考虑部分可观测性和物理约束。现有基准常忽略这些关键因素，难以评估任务的可行性与安全性。我们提出SPOC，一个面向安全感知的具身任务规划基准，集成严格的部分可观测性、物理约束、分步规划及基于目标的评估方式。涵盖火灾、液体、伤害、物品损坏和污染等多样化家庭隐患，通过状态和约束双重在线指标实现严格评估。对前沿大模型的实验表明，当前模型在隐式约束下难以保证安全规划。代码与数据集已公开于https://github.com/khm159/SPOC。
原文摘要 · Abstract (English)
Embodied Task Planning with large language models faces safety challenges in real-world environments, where partial observability and physical constraints must be respected. Existing benchmarks often overlook these critical factors, limiting their ability to evaluate both feasibility and safety. We introduce SPOC, a benchmark for safety-aware embodied task planning, which integrates strict partial observability, physical constraints, step-by-step planning, and goal-condition-based evaluation. Covering diverse household hazards such as fire, fluid, injury, object damage, and pollution, SPOC enables rigorous assessment through both state and constraint-based online metrics. Experiments with state-of-the-art LLMs reveal that current models struggle to ensure safety-aware planning, particularly under implicit constraints. Code and dataset are available at https://github.com/khm159/SPOC
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。