arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2311.04403cs.AI
收藏

以人为中心的规划

Human-Centered Planning

Meta;Reality Labs Research, Meta(Meta现实实验室研究)

Yuliang Li, Nitin Kamra, Ruta Desai, Alon Halevy

AI总结:

针对以人为中心的一日规划需求,该研究开发了带自我反思能力的LLM规划器LLMPlan和可转换文本约束的符号规划器SymPlan,实验显示LLMPlan约束满足能力接近符号规划器且用户满意度更高。

摘要:

大语言模型(LLMs)近来在结构化输出任务上取得了显著进展,例如编程、机器人规划和数据库查询。打造AI驱动的个人助理的愿景同样涉及生成结构化输出,比如一日规划或海外旅行规划。在此场景下,由于规划由人类执行,输出无需满足严格的句法约束。一个实用的助理还应当能够整合用户用自然语言指定的模糊约束,这使得LLMs成为规划任务的一个极具吸引力的选择。\n我们研究了一日规划问题。我们开发了一种基于LLM的规划器LLMPlan,扩展了对其输出进行自我反思的能力;还开发了一种符号规划器SymPlan,具备将文本约束转换为符号表示的能力。尽管没有正式的约束规范,我们发现LLMPlan平均而言实现了与传统符号规划器类似的显式约束满足(性能差异为2%),同时保留了对隐式需求的推理能力。因此,在40名用户参与的交互式评估中,基于LLM的规划器在用户满意度上优于对应的符号规划器(70.5% 对比 40.4%)。

英文摘要:

LLMs have recently made impressive inroads on tasks whose output is structured, such as coding, robotic planning and querying databases. The vision of creating AI-powered personal assistants also involves creating structured outputs, such as a plan for one's day, or for an overseas trip. Here, since the plan is executed by a human, the output doesn't have to satisfy strict syntactic constraints. A useful assistant should also be able to incorporate vague constraints specified by the user in natural language. This makes LLMs an attractive option for planning. We consider the problem of planning one's day. We develop an LLM-based planner (LLMPlan) extended with the ability to self-reflect on its output and a symbolic planner (SymPlan) with the ability to translate text constraints into a symbolic representation. Despite no formal specification of constraints, we find that LLMPlan performs explicit constraint satisfaction akin to the traditional symbolic planners on average (2% performance difference), while retaining the reasoning of implicit requirements. Consequently, LLM-based planners outperform their symbolic counterparts in user satisfaction (70.5% vs. 40.4%) during interactive evaluation with 40 users.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑