arXiv 2309.013522023-09-04cs.CLcs.AI

自驱动接地：具有自动语言对齐技能学习的大型语言模型智能体

Self-driven Grounding: Large Language Model Agents with Automatical Language-aligned Skill Learning

Intelligent Software Research Center, Institute of Software, CAS（中国科学院软件研究所智能软件研究中心）；SKL of Processors, Institute of Computing Technology, CAS（中国科学院计算技术研究所处理器国家重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）；Cambricon Technologies（寒武纪科技）；University of Science and Technology of China（中国科学技术大学）

Shaohui Peng, Xing Hu, Qi Yi, Rui Zhang, Jiaming Guo, Di Huang, Zikang Tian, Ruizhi Chen, Zidong Du, Qi Guo, Yunji Chen, Ling Li

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出自驱动接地框架，让LLM智能体通过自动提出并验证子目标来学习语言对齐技能，在BabyAI任务上以少量交互达到与模仿学习相当的性能。

摘要：

大型语言模型（LLMs）展示了其强大的自动推理和规划能力，并具备关于人类世界的丰富语义知识。然而，接地问题仍然阻碍了LLMs在真实世界环境中的应用。现有研究尝试微调LLM或利用预定义的行为API来桥接LLM与环境，这不仅为每个单一任务耗费大量人力定制，还削弱了LLMs的通用性优势。为了自主地将LLM接地到环境中，我们提出了自驱动接地（SDG）框架，通过自驱动的技能学习自动且渐进地实现LLM的接地。SDG首先利用LLM提出实现任务的子目标假设，然后通过与底层环境交互来验证假设的可行性。一旦验证成功，SDG便能在这些成功接地的子目标指导下学习泛化技能。这些技能可进一步用于完成更复杂的、未能通过验证阶段的任务。在著名的指令跟随任务集BabyAI上验证，SDG在最具挑战性的任务中取得了与耗费数百万演示的模仿学习方法相当的性能，证明了所学技能的有效性，并展示了我们框架的可行性和效率。

英文摘要：

Large language models (LLMs) show their powerful automatic reasoning and planning capability with a wealth of semantic knowledge about the human world. However, the grounding problem still hinders the applications of LLMs in the real-world environment. Existing studies try to fine-tune the LLM or utilize pre-defined behavior APIs to bridge the LLMs and the environment, which not only costs huge human efforts to customize for every single task but also weakens the generality strengths of LLMs. To autonomously ground the LLM onto the environment, we proposed the Self-Driven Grounding (SDG) framework to automatically and progressively ground the LLM with self-driven skill learning. SDG first employs the LLM to propose the hypothesis of sub-goals to achieve tasks and then verify the feasibility of the hypothesis via interacting with the underlying environment. Once verified, SDG can then learn generalized skills with the guidance of these successfully grounded subgoals. These skills can be further utilized to accomplish more complex tasks which fail to pass the verification phase. Verified in the famous instruction following task set-BabyAI, SDG achieves comparable performance in the most challenging tasks compared with imitation learning methods that cost millions of demonstrations, proving the effectiveness of learned skills and showing the feasibility and efficiency of our framework.

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。

AI总结：

摘要：

英文摘要：

相关论文