arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2406.02791cs.AIcs.CLcs.RO
收藏

语言模型可以从环境反馈中推断符号规划器的动作语义

Language Models can Infer Action Semantics for Symbolic Planners from Environment Feedback

University of Southern California(南加州大学)

Wang Zhu, Ishika Singh, Robin Jia, Jesse Thomason

AI总结:

本文提出PSALM,结合符号规划器与LLM,通过环境反馈自动学习动作语义,将计划成功率从36.4%提升至100%。

摘要:

符号规划器在给定专家定义的、特定领域的逻辑动作语义后,能够发现从初始状态到目标状态的动作序列。大型语言模型(LLMs)可以直接生成此类序列,但其在推理和状态跟踪方面的局限性往往导致生成的计划不充分或不可执行。我们提出了语言模型预测动作语义(PSALM),该方法通过结合符号规划器和LLMs的优势,自动学习动作语义。PSALM反复提出并执行计划,利用LLM部分生成计划,并根据执行结果推断特定领域的动作语义。PSALM维护对可能动作语义的信念,该信念被迭代更新,直到达到目标状态。在7个环境上的实验表明,仅从一个目标学习时,PSALM将计划成功率从36.4%(在Claude-3.5上)提升至100%,并且比先前工作更高效地探索环境,以推断真实领域的动作语义。

英文摘要:

Symbolic planners can discover a sequence of actions from initial to goal states given expert-defined, domain-specific logical action semantics. Large Language Models (LLMs) can directly generate such sequences, but limitations in reasoning and state-tracking often result in plans that are insufficient or unexecutable. We propose Predicting Semantics of Actions with Language Models (PSALM), which automatically learns action semantics by leveraging the strengths of both symbolic planners and LLMs. PSALM repeatedly proposes and executes plans, using the LLM to partially generate plans and to infer domain-specific action semantics based on execution outcomes. PSALM maintains a belief over possible action semantics that is iteratively updated until a goal state is reached. Experiments on 7 environments show that when learning just from one goal, PSALM boosts plan success rate from 36.4% (on Claude-3.5) to 100%, and explores the environment more efficiently than prior work to infer ground truth domain action semantics.

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。

↑